Abstract

From a series of 282 consecutive temporal resections for medically intractable epilepsy associated with mesial temporal sclerosis (MTS), dysembryoplastic neuroepithelial tumour (DNT) or non-specific pathology (NSP), 51 patients had persistent or recurrent seizures occurring at least monthly. Of these patients, 44 underwent detailed assessment of their postoperative seizures, which included clinical evaluation, interictal and ictal EEG and high-resolution MRI. Of the 20 patients with MTS in the original pathology, 14 (70%) had postoperative seizures arising in the hemisphere of the resection, the majority (12 patients) in the temporal region. Although MRI demonstrated residual hippocampus in five of these 12 patients, only one patient was considered to have seizures arising there, whilst the remainder had electroclinical evidence of seizure onset in the neocortex. In contrast, five of the MTS relapses (25%) had seizure onset exclusively in the contralateral temporal region. Among the 14 patients with non-specific pathology, relapse was also predominantly from the ipsilateral hemisphere (64%), but more relapsed from extratemporal sites compared with the MTS cases, including two with NSP who had occipital structural abnormalities. Although 70% of the 10 patients with DNT had postoperative partial seizures arising in the ipsilateral hemisphere, many (60%) had evidence of a more diffuse disorder with additional generalized seizures, cognitive and behavioural disturbance and multifocal and generalized EEG abnormalities. Nine patients (20%) had immediate postoperative seizure-free periods of at least 1 year, and seven of these had MTS in the operative specimen. Of these seven patients, four had ipsilateral temporal seizures and three had contralateral temporal seizures. Overall, few missed lesions were discovered on postoperative MRI and reoperations were performed or considered possible in a minority of cases. Despite well-defined preoperative electroclinical syndromes of temporal lobe epilepsy, many patients relapsed unexpectedly, either immediately or remotely from the time of surgery. Maturing epileptogenicity in a surgical scar was not, however, considered to be a significant primary mechanism in patients who relapsed after a seizure-free interval.

Introduction

The proportion of patients becoming seizure-free after temporal lobe resection for medically intractable epilepsy has been estimated to range from 55 to 70% (Elwes et al., 1991; Engel et al., 1993). Several reports (Wyler et al., 1989; Awad et al., 1991; Germano et al., 1994) have described the outcome of patients who are reoperated and report short- to medium-term seizure remissions of 50–65%. From the Montreal series of 559 cases of anteromedial temporal lobectomy performed between 1971 and 1990, 40 patients with persistent seizures had further surgery to the temporal lobe (Germano et al., 1994). MRI before the second procedure identified residual mesiotemporal structures in all patients, EEG abnormalities being confined to the ipsilateral temporal region in 32 of the 40 cases. After further hippocampal resection, combined with additional neocortical resection in eight patients, 65% of the patients became seizure-free. Those patients with a poor outcome had EEG abnormalities extending beyond the temporal region, implying that the epileptogenic zone was not confined to residual mesial temporal structures. No failure was due to non-identification of a structural lesion in the temporal lobe. From an operated series of 154 resections for intractable epilepsy, Awad and colleagues reported that 72 patients had persistent seizures; 15 of these patients were finally selected and reoperated (Awad et al., 1991). Eleven of these had undergone temporal resection initially. The second procedures in these 11 cases included resection of residual mesial temporal structures in six; lateral temporal resections in two who had previous extensive resection of the mesial temporal structures; and more extensive temporal resections that included an incompletely removed foreign tissue lesion in three (hamartoma, vascular hamartoma and vascular malformation). In two of these the lesion had not been detected before the first operation. All resections were performed after the seizure had been localized with invasive electrodes, and complete seizure control occurred in 64% of patients. Wyler and colleagues reported the investigation and outcome of 37 patients selected from a series of 319 operated cases (Wyler et al., 1989). Second resections were temporal in 23, with a seizure-free outcome in 52% of the patients. All patients who had a seizure-free outcome had mesial temporal sclerosis (MTS) in the original resection, and repeat surgery consisted of further hippocampal resection. Seven patients had structural lesions that had not been removed by the original procedure. Four of these cases involved second operations on the frontal lobe, two cases involved temporal operations after initial frontal resections, and only one case involved resection of a temporal lesion on the side where the previous temporal resection had failed. These reports have emphasized the inadequacy of the first surgical resection (especially the retention of mesial temporal structures and the incomplete removal of structural lesions) as a major factor in continuing postoperative seizures.

There are, however, other possible mechanisms of continuing seizures. In centres that employ more aggressive hippocampal resection combined with anteromedial lobectomy (Spencer et al., 1984; Polkey, 1988), the problem of inadequate mesial resection may be less important. In cases of persistent seizures after such resections, it is possible that the epileptogenic zone is more extensive, involving the posterolateral temporal neocortex or other contiguous extrahippocampal tissue. This may seem likely in patients in whom pathological examination of excised mesial temporal structures reveals only minor non-specific abnormalities. However, the role of persisting epileptogenicity in the vicinity of the temporal lobectomy site after apparently complete resection of a sclerotic hippocampus has not been established. Other mechanisms could include relapse from the contralateral temporal lobe, perhaps in the context of bilateral mesial temporal sclerosis or by means of a mirror focus in the context of a unilateral low-grade tumour. The late recurrence of seizures could suggest maturing epileptogenicity in a surgical scar.

Advances in neuroimaging have highlighted the problem of dual pathology where either temporal or extratemporal structural abnormalities coexist with MRI evidence of hippocampal sclerosis. The frequency of this association in quantitative MRI studies of refractory partial epilepsy is ~15% (Cascino et al., 1993; Raymond et al., 1994a; Cendes et al., 1995). Often, these lesions are subtle and are part of the spectrum of a neuronal migration disorder. In these cases the optimum surgical strategy appears to be the removal of both the lesion and the atrophic hippocampus (Li et al., 1999); thus, the removal of an atrophic hippocampus alone might explain surgical failure.

Whatever the explanation, it appears that mechanisms of relapse are probably heterogeneous. With the growth of temporal lobe epilepsy surgery, such cases will inevitably form a significant group of patients with intractable epilepsy. In the light of the reported benefits of reoperation, it has been the practice of our department in recent years to offer reassessment to patients with persistent or recurrent disabling seizures. Investigations are pursued until a reasonable electroclinical classification of the seizure syndrome is obtained using clinical, imaging and neurophysiological evidence. To our knowledge, the relative importance of the proposed mechanisms of failure outlined above has not been established in unselected surgical failures after temporal resection. The aim of the present study is, therefore, to describe the electroclinical patterns of seizure recurrence in such a population and, where possible, to infer an anatomical and functional basis for these seizures.

Patients and methods

Patient selection

From a series of 282 consecutive temporal resections for intractable epilepsy associated with MTS (n = 165), dysembryoplastic neuroepithelial tumour (DNT) (n = 77) and normal or non-specific pathology (NSP) (n = 40), 56 (20%) had persistent or recurrent postoperative seizures occurring at least monthly. All operations had been performed by the same neurosurgeon (C.E.P.) between 1976 and July 1995. Two hundred and thirty-five patients had temporal en bloc resection (Polkey, 1988). A further 47 selective amygdalo-hippocampectomies (Yasargil et al., 1985) were performed in which the resection was limited to the mesial structures, including most of the parahippocampal gyrus.

Forty-four of the 56 patients (79%) finally underwent reassessment (MTS, n = 20; NSP, n = 14; DNT, n = 10). In the remaining 12 patients reassessment was not possible because of foreign residence (n = 4), loss to follow-up (n = 3), death (n = 2), seizure-freedom attained during the period of reassessment (n = 1), failure to establish an electroclinical classification (n = 1) and patient refusal (n = 1).

Preoperative assessment

The present cohort of patients was assessed and operated over a long period (1980–95) and therefore preoperative assessment was not uniform. In general, operations were performed if there were congruent data from the clinical history and the neuropsychological, neurophysiological and neuroimaging investigations. Emphasis on any one modality of investigation depended on the details of the individual patient and on the suspected underlying pathology. The preoperative details of patients with MTS and NSP are set out in Tables 1 (MTS) and 2 (NSP) and similar details of patients with DNT are included in Table 4. It is clear that individual circumstances dictated the reason for operation in some cases, and it is difficult to document precisely the clinical motivation by retrospective assessment. Furthermore, it was difficult to appreciate in the assessment of the preoperative data the expectation of seizure outcome after surgery. For example, the demonstration of a temporal lobe mass lesion in a child with intractable seizures despite imperfect electroclinical correlation was the prompting factor for surgery in three patients with temporal lobe DNT. Overall, however, patients were operated on the basis of reasonably convincing data and, whatever the expectation for freedom from seizures was before surgery, our major concern was to establish the electroclinical patterns of postoperative seizure relapse and how that might influence our approach to similar patients in the future.

Postoperative assessment

Neuropathology

The neuropathological reports on the excised specimens were reviewed. Over the years, the temporal lobe surgical specimens had been reported systematically by experts in the surgical pathology of epilepsy (Honavar et al., 1997). MTS was considered to be present if the original report had specified neuronal loss and gliosis in the hippocampus, with the most severe changes in the Sommer sector (H1) and the end-folium (H3–5) and less severe changes in the dentate fascia and the resistant zone (H2) (Margerison and Corsellis, 1966; Bruton, 1988). In many cases the sclerotic process involved other mesial structures, such as the parahippocampal gyrus, amygdala and uncus, and MTS is therefore the preferred term in this article. Minor degrees of white matter neuronal ectopia, although always noted, were not considered to represent dual pathology. Quantitative techniques were not employed. The original slides of patients with low-grade tumours were reviewed because of the emergence of DNT as a diagnostic entity in the late 1980s (Daumas-Duport et al., 1988). In addition, the completeness of histological resection was clarified as this was not always stated in the original report. The pathology was deemed non-specific if no structural lesion was present and the classic appearances of MTS were absent. Minor degrees of gliosis confined to the end-folium were also considered non-specific and likely to be related to the consequences of chronic seizure activity rather than a primary epileptogenic lesion (Sagar et al., 1987).

Clinical review

All patients and/or their carers underwent a clinical interview with regard to the details of their postoperative seizures. Emphasis was placed on the ictal semiology of recorded seizures on video-telemetry, but in the small number of cases in which seizure recording did not prove possible, certain clinical features were regarded as reliable, in terms of lateralization, from the clinical history. These included hemisensory or motor phenomena, ictal speech or speech arrest and postictal dysphasia. Version of the head and limb automatisms, however, were not considered reliable from the history alone. The following ictal features observed on telemetry were considered to have lateralizing significance with regard to the seizure focus: (i) version of the head and eyes: early non-forced version of the head and eyes, ipsilateral; late forced version, often before generalization, contralateral (Wyllie et al., 1986; Kotagal, 1991; Jayakar et al., 1992; Williamson et al., 1998); (ii) unilateral dystonic posturing, usually of an upper limb, contralateral (Kotagal et al., 1989; Newton et al., 1992; Fakhoury et al., 1995; Williamson et al., 1998); (iii) unilateral automatisms, usually of the hand (usually associated with contralateral dystonia), ipsilateral (Kotagal et al., 1989; Thadani et al., 1995); (iv) speech: ictal speech, non-dominant hemisphere; speech arrest, postictal dysphasia, dominant hemisphere (Koerner et al., 1988; Privitera et al., 1991; Fakhoury et al., 1994; Yen et al., 1996; Williamson et al., 1998).

Early forced contralateral version followed by clonic activity of the contralateral limbs and generalization was considered to be compatible with seizure onset in the posterior quadrant, the prominent motor semiology probably representing rapid spread into the frontal region (Ludwig et al., 1975; Duchowny et al., 1994). Acoustic auras, i.e. elemental sounds consisting of simple noises, piercing or buzzing, were considered to indicate seizure origin from the lateral temporal neocortex (Commission on Classification and Terminology of the International League Against Epilepsy, 1985).

Neurophysiology

All patients had routine and sleep interictal EEG recordings. The majority of patients had continuous video-EEG monitoring with scalp electrodes. The topography of ictal scalp recording (i.e. midtemporal–sylvian versus anterior temporal maximum) was not considered to distinguish residual mesial from lateral temporal onset reliably, as we were uncertain how these patterns could be influenced in the postlobectomy state. Furthermore, it appears that the topography of temporal spikes is probably determined more by the conductivity of the skull and its foramina than by the distribution of discharges over the surface of the brain (Fernandez Torre et al., 1999). Because of this, lateralized ictal onset on scalp EEG was considered reliable in the postoperative setting. Furthermore, we felt that rapid contralateral spread might be less likely owing to disruption of the interhippocampal commissures (Spencer et al., 1987). Some patients had additional intracranial EEG studies with foramen ovale, subdural strip or grid recordings but, reflecting the routine clinical practice of the reassessment process, these studies were undertaken only in those patients in whom a further surgical procedure was considered possible.

Neuroimaging

MRI was performed using a 1.5 tesla GE Signa Horizon unit (GE Medical Systems, Milwaukee, Wis., USA). Coronal T2-weighted fast spin echo images [repetition time/echo time (TR/TEeff) 4300/84 ms, echo train length 14, field of view (FOV) 22 × 16.5 cm, matrix 256 × 192, slice thickness 3.5 mm, gap 0.5 mm] were obtained perpendicular to the long axis of the temporal lobe. T1-weighted 3D SPGR (spoiled grass) images of the whole brain (flip angle 35°, TR/TE 14/3 ms, FOV 22 × 16.5 cm, matrix 256 × 192) were acquired to give 124 contiguous 1.5 mm coronal partitions. In an attempt to visualize the hippocampus more clearly, fluid-attenuated inversion recovery (FLAIR) sequences were obtained in patients whose electroclinical syndrome suggested onset in the contralateral temporal lobe. This sequence was not obtained systematically in other cases. The following aspects were reviewed on all scans: the extent of hippocampal resection; the presence of identifiable amygdala; the presence of periresection gliosis or encephalomalacia; the appearance of the contralateral temporal lobe; and the possibility of missed lesions, such as small tumours or a malformation of cortical development in the neocortex or periventricular heterotopia. We chose to regard a hippocampal resection as complete if, on visual inspection of the MRI, the resection extended to the level of the middle of the midbrain posteriorly, at least to the level of the posterior margin of the crus cerebri. Resection to this level has been associated with seizure-free outcomes in two detailed studies of outcome related to the extent of hippocampal resection, as judged by postoperative MRI (Nayel et al., 1991; Wyler et al., 1995). However, preoperative quantitative MRI assessment of unilateral hippocampal atrophy shows this process to be most commonly diffuse (Van Paesschen et al., 1997) and we cannot discount with certainty seizure onset from minimal amounts of retained posterior hippocampus. On a practical level, the possibility of a further hippocampal resection was considered for each case and the clinical judgement of no remaining hippocampal surgical target correlated in every case with a resection that extended to this anatomical landmark.

Results

General characteristics of the re-evaluated group

Of the 44 patients with postoperative seizures, 23 were male and 21 were female. Mean age at original surgery was 26 years (range 4–59 years) and the mean follow-up interval from surgery to reassessment was 6 years (range 3–17 years).

Postoperative electroclinical classification

The pre- and postoperative seizure details and the results of EEG and neuroimaging, together with the electroclinical classification for each patient, are given in Tables 3 (MTS and NSP) and 4 (DNT). A summary of the electroclinical classification for the overall group and for each of the pathological subgroups is given in Table 5. Thirty patients (68%) had at least one seizure type arising in the hemisphere ipsilateral to the original resection. Twenty patients (45%) had persistent/recurrent seizures classified as ipsilateral temporal, including three (Cases 11, 19 and 20) who had an additional seizure focus. Nine patients (20%) had at least one contralateral seizure but one of these was of frontal origin (Case 19) and two (Cases 20 and 38) also had ipsilateral seizures.

Mesial temporal sclerosis

Ipsilateral seizures and residual hippocampus

Fourteen patients had at least one seizure arising in the hemisphere ipsilateral to the resection. Twelve of these were classified as ipsilateral temporal. Although MRI imaging revealed residual hippocampus in five patients (Cases 4, 8, 12, 19 and 20), the electroclinical features were supportive of seizures arising in this remnant in only two (Cases 8 and 19). A further hippocampal resection was performed in the first patient (Case 8), and this patient was seizure-free at the 6 month follow-up. In the other patient (Case 19), although one seizure was related to the residual hippocampus on intracranial EEG recording, the major (new) seizure type arose from the contralateral frontal lobe.

In the remaining three patients with residual hippocampus, seizures with a buzzing aura were localized in one (Case 4) to the posterior temporal–sylvian region on intracranial mat electrode recording; no ictal activity was recorded by a depth electrode in the remaining hippocampus. In the second patient (Case 12), postoperative seizures were considered to be of ipsilateral frontal origin and the third patient (Case 20) had persistent ipsilateral temporal seizures with an acoustic aura and an additional seizure type of contralateral temporal origin.

Ipsilateral seizures and complete hippocampal resection

The remaining nine patients with MTS and ipsilateral seizures had complete hippocampal resections. Seizure semiology remained unchanged in one (Case 9). Another patient (Case 1) lost a preoperative aura of scrotal sensation and fear but continued to have seizures of temporal lobe semiology. In three patients (Cases 2, 6 and 11), epigastric auras were replaced by a non-specific aura in one and acoustic auras in two. In one of these patients (Case 11), another seizure type of ipsilateral frontal location also developed. Another patient (Case 3), without a preoperative aura, also developed a new seizure type with an acoustic aura.

Two patients had preoperative imaging evidence of more diffuse extrahippocampal atrophy. In one (Case 10), complex partial seizures of the mesial temporal type associated with an olfactory aura, localized by depth recording to the right amygdala, were relieved but generalized seizures, which were heralded by hemiclonic jerking and had a more diffuse temporal onset on the preoperative intracranial evaluation, persisted. The second patient (Case 5) relapsed after 14 months with similar complex partial seizures but with prominent Todd's paralysis. Interictal EEG discharges were localized to the region of the resection, but seizure recording was not possible because of a severe behavioural disturbance.

The final patient (Case 7) was free of complex partial seizures of left mesial temporal origin for 8 months but then relapsed with a new partial seizure of speech arrest and generalized seizures with postictal dysphasia.

Coexistence of partial and cryptogenic generalized epilepsy

One patient (Case 18) with mild learning disability had relief of left temporal complex partial seizures but continued to experience more frequent generalized tonic–clonic seizures associated with polyspike and wave discharges on a slow EEG background. In addition, frequent runs of fast positive spikes were present in the mid-parietal regions bilaterally, which was considered to indicate some underlying distortion of the cortical gyri (Matsuo and Knott, 1977). MRI showed no cortical abnormalities and this patient was considered to have a coexistent persistent cryptogenic generalized epilepsy.

Contralateral seizures

Seven patients with MTS had contralateral seizures but two of these (Cases 19 and 20) also had seizures localized to the temporal region ipsilateral to the resection. In Case 19, the contralateral seizure (which emerged after 2 years of postoperative freedom from seizures) was classified as frontal with an apparently normal contralateral hippocampus on MRI. In the five patients (Cases 13–17) with exclusively contralateral temporal seizures, MRI showed signal changes compatible with sclerosis in three and was normal in one. The fifth patient died before an MRI was obtained. In these patients, there was no evidence of contralateral epileptogenesis in the preoperative evaluation. Four of the five patients had preoperative seizure recording, three with foramen ovale electrodes. None had contralateral seizures recorded. The patient without preoperative seizure recording, however, remained seizure-free for 4 years before relapse. Similarly, in patient 16, contralateral seizures developed after 4 years of complete freedom from seizures.

Non-specific pathology

In the 14 patients with non-specific pathological findings, the results of the electroclinical classification and postoperative imaging were more diverse.

Ipsilateral temporal seizures

Four patients had ipsilateral temporal seizures. One (Case 24) was considered to have seizures arising in a substantial hippocampal remnant seen on MRI, but further surgery was considered inappropriate because of failure of the contralateral hemisphere to support memory adequately on Sodium Amytal testing. In the remaining three patients, the clinical features were suggestive of extrahippocampal seizures, presumably arising from a site surrounding the previous resection. In Case 21, new seizures, consisting of staring and early slow version of the head followed by flexion of both arms, facial grimacing and hemiclonic jerking (replacing seizures of mesial temporal semiology of hippocampal origin on depth recording), were considered to arise more probably from the posterior temporal neocortex with rapid extratemporal propagation than from a small posterior hippocampal remnant. In Case 22, preoperative seizures consisting of buzzing sounds had progressed to staring, oral automatisms and contralateral dystonia but progressed postoperatively to numbness of the tongue and palate, blinking, hemifacial jerking, and numbness of the contralateral arm and trunk, sometimes leading to generalization. In Case 23, pre- and postoperative seizures were characterized by initial speech arrest. Although preoperative foramen ovale seizure recordings were localized to the left temporal region, the initial discharges were of greater amplitude in the more superficial electrode contacts, suggesting neocortical origin.

Two other patients had persisting seizures with auras of buzzing sounds, but the postoperative ictal-EEG was poorly localized in the hemisphere of the resection in one (Case 25) and non-lateralizing in the other (Case 26). Again, review of the preoperative foramen ovale electrode recordings showed a pattern suggestive of neocortical origin with either less prominent or late involvement of the deep electrode contacts in close proximity to the mesial temporal structures.

Ipsilateral temporal replaced by ipsilateral extratemporal seizures

In Case 28, preoperative musicogenic partial seizures were replaced by hemisensory attacks localized to the hand area of the sensory cortex by intracranial EEG studies. A similar pattern was observed in Case 27 (complete hippocampal resection), in which preoperative complex partial seizures with temporal semiology were replaced by nocturnal seizures with EEG localization to the ipsilateral frontal lobe.

Outcome of temporal resection in the setting of extratemporal structural lesions

Three patients had extratemporal mass lesions. All had been recognized preoperatively, and temporal resection was performed on the basis of electroclinical localization with depth electrodes in two patients with occipital lesions and with foramen ovale electrode recording in a patient with a hypothalamic hamartoma (Case 32). In this case, the habitual temporal lobe seizure type was alleviated but gelastic and generalized seizures continued and new ipsilateral suprasylvian seizures with hemisensory symptoms emerged. This patient, in the absence of demonstrated seizure origin within the hamartoma, was considered to have a multifocal epilepsy syndrome, the resection serving to alter propagation patterns. The two patients with occipital lesions had initially favourable outcomes from surgery, with 2 years of seizure-freedom before relapse in one (Case 30) and infrequent nocturnal generalized seizures for the first 6 postoperative years in the other (Case 29).

Contralateral seizures

Only one patient (Case 31) with non-specific pathology had evidence of contralateral temporal seizures, preoperative seizures having been localized to the operated temporal lobe on depth recording without evidence of additional contralateral seizures.

Non-epileptic seizures

Two patients (Cases 33 and 34) had preoperative clinical and ictal-EEG localization to the operated temporal lobe but continued to have atypical attacks postoperatively, which proved to be non-epileptic on ictal recording. No psychiatric predisposing factors were identified in these patients.

Dysembryoplastic neuroepithelial tumour

Of the 10 patients with pathological evidence of DNT in the operative specimen, seven had habitual seizures localized to the ipsilateral hemisphere.

Ipsilateral temporal seizures

Four patients had persistent seizures in the temporal region. In Case 44, the seizure consisted of a persisting acoustic aura leading to aphasia and generalization. MRI showed residual tumour with diffuse infiltration of the remaining left superior temporal gyrus. In the other three patients, the seizure focus was regarded as predominantly posterior temporal in location. Two of these patients (Cases 35 and 36) had initial staring followed by rapid version of the head and posturing or clonic activity of the contralateral limbs. In Case 35, this replaced the preoperative seizure of staring, automatisms and version localized to the mesial temporal region by foramen ovale recordings. However, preoperative CT had demonstrated a predominantly posterior temporal tumour with extension to the hippocampus. Although the postoperative MRI showed no evidence of residual tumour, histological review suggested incomplete resection. In Case 36, the preoperative CT was regarded as normal and the operation had been performed on the basis of electroclinical localization to the temporal lobe. The postoperative MRI, however, showed an incompletely resected posterior temporal tumour. This was the only example in which a lesion was discovered adjacent to the resection posteriorly. Intracranial EEG recording confirmed an inferobasal temporal onset of seizures. This lesion was resected and confirmed as a DNT and the patient was seizure-free 4 months after surgery. The third case differs in that the preoperative CT showed an anterior temporal lesion. This patient had two seizure types, one of which was abolished by the resection (fear, aphasia, dystonia and version), the second persisting unchanged and consisting of nocturnal tonic–clonic seizures, preceded by version and unilateral limb-jerking. The postoperative MRI suggested complete tumour resection, but histological review of the operative specimen suggested incomplete tumour resection.

Ipsilateral frontal seizures

Postoperative seizures were localized to the frontal region in three patients. In two (Cases 38 and 42), a more diffuse abnormality was evident in the form of severe behavioural and developmental disturbance, together with multifocal and generalized epileptiform abnormalities on EEG. In both patients, incomplete lesion resection was evident on histology and in one (Case 38) the tumour was associated with a band of cortical dysplasia which extended anteriorly where, postoperatively, seizure onset was demonstrated in the frontal region by intracranial EEG. The third patient (Case 43) had a lesser degree of learning disability with multifocal epileptiform abnormalities on EEG. Despite histological and MRI evidence of complete tumour removal, postoperative partial seizures were localized clinically and by scalp EEG to the ipsilateral frontal region. Isolated generalized seizures and drop attacks, described in the history, were not telemetered. All three patients had relief of temporal lobe complex partial seizures after surgery.

DNT in the setting of symptomatic generalized epilepsy

The most frequent pathological diagnosis in patients classified as having a generalized/multifocal seizure disorder was DNT. In three patients (Cases 39, 40 and 41) multiple or generalized seizure types, intellectual regression, behavioural disturbance and generalized and multifocal EEG abnormalities persisted after surgery. In none of these patients could the clinical picture be clearly attributed to the effects of a localized temporal abnormality or to the effects of chronic epilepsy as all of these features were evident from the onset of the seizure disorder. As described above, three further cases had more clearly defined postoperative partial seizures (and remission of preoperative temporal lobe seizures) but had, in addition, widespread EEG abnormalities and cognitive and behavioural problems. Although neuroimaging demonstrated residual tumour in five and histological review of the operative specimen demonstrated tumour adjacent to the resection margin in eight of the cases with DNT, this feature did not aid significantly in the electroclinical classification of cases with postoperative generalized or multifocal syndromes.

Seizure relapse after an initial seizure-free period

Nine patients (20%) (Cases 1, 5, 8, 11, 15, 16, 19, 23 and 30) had immediate postoperative seizure-free periods of at least 12 months. The mean duration of postoperative freedom from seizures for these patients was 38 months. All cases relapsed to having at least monthly seizures and all proved eventually to be medically refractory (all had at least 2 years of follow-up from the time of relapse). Seven had MTS in the operative specimen, whereas only two patients with NSP pathology and none with DNT relapsed after an initial prolonged period of freedom from seizures after surgery. In MTS patients, four relapsed from an ipsilateral temporal site, of whom three had a complete hippocampal resection. Three of the seven cases of MTS with contralateral seizures were also in this group. Three other patients with MTS had postoperative seizure-free periods ranging from 5 to 8 months. One patient (Case 13) had contralateral temporal and two (Cases 6 and 7) had ipsilateral temporal seizures, both with complete hippocampal resections.

Discussion

In this paper, we have proposed an electroclinical classification of persistent seizures after temporal lobe epilepsy surgery. In patients with mesial temporal sclerosis, relapse came from the hemisphere of the temporal resection in 70% of cases (55% from the ipsilateral temporal region) and from the contralateral hemisphere in 35% (30% from the contralateral temporal region). Patients with non-specific pathology also relapsed predominantly from the ipsilateral hemisphere (64%), but these cases were more heterogeneous in that more relapsed from extratemporal sites on that side than in cases of MTS. Although 70% of cases of DNT had postoperative partial seizures localized to the hemisphere of the resection, many also had evidence of a more diffuse disorder with additional generalized seizures, EEG abnormalities and cognitive and behavioural impairment, which in some cases amounted to an epileptic encephalopathy. Overall, only one structural lesion was discovered on postoperative MRI and no dual pathology was diagnosed on imaging of MTS relapses. The majority of seizures in patients with MTS or NSP pathology were regarded as extrahippocampal and reoperations were performed or considered possible in a small number of patients. Many patients had a well-defined preoperative electroclinical syndrome of temporal lobe epilepsy, only to relapse unexpectedly, either immediately or remotely from the time of surgery.

Unlike other authors describing reoperation after surgical resection (Wyler et al., 1989; Awad et al., 1991; Germano et al., 1994), we concluded that most patients with MTS did not have recurrent seizures arising in residual hippocampus. This conclusion was based both on the demonstration of a completely resected hippocampus and on the electroclinical characteristics of the seizures, which were suggestive of neocortical onset. We also considered the possibility of seizures arising in residual amygdala. Bruton described neuronal loss and gliosis in 76% of amygdalae removed in conjunction with Ammon's horn sclerosis (Bruton, 1988), and we confirmed a high incidence of pathological involvement of the amygdala, usually of a non-specific nature, in cases that relapsed from an ipsilateral site. Amygdalar tissue was described in every pathological specimen from these cases and, although it is likely that small amounts of amygdala remain after surgery, this seems an unlikely source of residual seizures, particularly as efferent pathways (Amaral et al., 1986) are likely to have been disconnected. The most important clinical characteristic supporting an extrahippocampal origin of recurrent seizures in the MTS group was the presence of acoustic auras. Five patients had acoustic auras, occurring de novo in three, all with complete hippocampal resections. Penfield and Jasper demonstrated that elemental auditory responses could only be obtained by stimulation of the posterior third of the sylvian fissure, Brodmann areas 41 and 42. Rarely, when the upper bank of the fissure had been removed, it was possible to demonstrate that the response extended inwards in the first temporal convolution and corresponded with the transverse gyrus of Heschl (Penfield and Jasper, 1954). We confirmed, on intracranial recording, seizure onset in the posterior-temporal/sylvian region (Case 4) without involvement of a substantial hippocampal remnant. Auditory auras are, however, infrequently described in temporal lobe epilepsy surgery series (Gloor, 1990; French et al., 1993) and we are aware of no previous reports of this pattern of seizure relapse after surgery for MTS.

Other features supporting extrahippocampal seizure onset included clinical and ictal EEG support for de novo postoperative frontal lobe seizures, and in two patients with complete hippocampal resections seizures began with initial speech arrest in one and focal jerking of the contralateral limbs in another. In addition, two cases had preoperative MRI evidence of more widespread extrahippocampal atrophy. Such a finding is not new, having been first observed by Falconer, who noted that, in cases of mesial temporal sclerosis, the hemicranium and in particular the middle cranial fossa were often smaller radiologically on the affected side (Falconer, 1964). Similarly, neuropatholgical studies have shown that, whilst the maximum neuronal loss and scarring is in the hippocampus, the process may extend to involve the parahippocampal gyrus, the amygdala and, in some cases, the uncus and the superior temporal gyrus as far forward as the anterior part of the sylvian fissure (Margerison and Corsellis, 1966). Bruton observed that all cases of temporal resection with mesial temporal sclerosis had sclerotic changes beyond the confines of the hippocampus, mostly consisting of gliosis in the adjacent temporal lobe white matter but also involving cortical neuronal damage in the temporal gyri in 27% of cases (Bruton, 1988).

Similarly, quantitative MRI studies have demonstrated a range of unilateral ipsilateral hemispheric, bilateral cortical and temporal lobe volume reductions in temporal lobe epilepsy (Jack et al., 1990; Lencz et al., 1992; Lee et al., 1995; Marsh et al., 1997), although these findings were not correlated with surgical outcome. Sisodiya and colleagues, however, using quantitative post-processing of preoperative MRIs in patients with MTS, described abnormal extrahippocampal structural changes in 14 patients who had complete hippocampal resections, 10 of whom did not become seizure-free. In contrast, 11 of 13 cases without these changes were seizure-free. No clinical details regarding the electroclinical features of postoperative seizures in these patients were provided but criteria similar to those used in the present study regarding the extent of hippocampal resection were employed (Sisodiya et al., 1997).

In the present patients, it is uncertain whether the extrahippocampal seizures after resection of MTS are dysgenetic or due to an initial brain insult that maximally affected the hippocampus, or reflect changes induced by seizure activity. Hippocampal sclerosis can occur with temporal lobe developmental lesions and in association with both temporal and extratemporal cortical dysplasia, most notably subependymal heterotopia (Kuzniecky, 1994; Raymond et al., 1994a; Cendes et al., 1995). In our series, postoperative high-resolution MRI failed to detect either gross or subtle dual pathology in the MTS failures, but the changes of dysplasia are not always visible on neuroimaging. Similarly, in the case of periventricular heterotopia, well-localized preoperative syndromes of mesial temporal epilepsy are described with poor outcome (Li et al., 1997), emphasizing that epileptogenic abnormalities may occur at a distance from the primary pathology. In the present patients, in whom relapse occurred from areas such as the temporal neocortex and the ipsilateral and contralateral frontal lobes, it seems probable that epileptogenesis is related to occult cortical dysplasia in these locations.

Relapse from the contralateral temporal lobe in MTS appears an intuitive explanation for surgical failure, particularly in view of autopsy studies of epileptic patients which report hippocampal sclerosis as being commonly bilateral with rates of bilateral pathology ranging from 47 to 86% (Sano and Malamud, 1953; Pfeiffer et al., 1963; Margerison and Corsellis, 1966; Meencke et al., 1991). In our series, although contralateral relapse occurred predominantly in MTS cases, this mechanism accounted for only 20% of cases overall. It is probable that preoperative assessment is most successful in precluding the majority of cases with bilateral temporal lobe epilepsy from surgery, and it is noteworthy that the patients in the series of Margerison and Corsellis had been assessed and specifically deemed inoperable. Although bilateral hippocampal atrophy has been found on high-resolution MRI scans in 9% (King et al., 1995) and 18% (Quigg et al., 1997) of operated patients, it appears that a successful outcome can be obtained if seizures are demonstrated to arise predominantly from one side. Our results suggest that relapse from the contralateral temporal lobe may occur many years after surgery and suggest that, when developing epileptogenesis occurs in association with bilateral hippocampal sclerosis, the process can be remarkably asymmetrical.

Although previous studies (Falconer et al., 1964; Duncan et al., 1987; Bruton, 1988; Nakasato et al., 1992; Zentner et al., 1995; Hennessy et al., 1999) have reported less favourable outcomes for temporal resections associated with normal or non-specific pathology, no detailed explanation is available as to why these cases fail. It is clear, on review, that four patients with acoustic auras and speech arrest and ictal EEG onsets suggestive of neocortical temporal origin had lateral temporal lobe epilepsy from the outset. In the two patients whose mesial temporal seizures were replaced by extratemporal seizures, the findings suggest the existence of a regional epileptogenicity in which the hippocampus represents the area of cortex with the lowest threshold for seizure generation, other areas of cortex with a higher threshold for seizure generation becoming the site of ictal onset after its removal. Furthermore, patterns of seizure propagation may be altered in some patients, in whom preoperative seizures of predominantly mesial temporal semiology are replaced by extratemporal seizures, usually involving secondary generalization. This pattern of relapse may be potentially deleterious for these patients, exposing them to the risk of tonic–clonic seizures, including sudden death (Hennessy et al., 1999). It is of interest that two patients in the NSP group became apparently seizure-free after surgery but quickly relapsed with non-epileptic seizures. Bruton similarly reported, from the earlier Maudsley series, that even though some patients with normal or indeterminate findings at pathology became seizure-free, none benefited psychosocially (Bruton, 1988). The underlying pathophysiology of intractable focal temporal lobe epilepsy associated with non-specific pathology is obscure. It is apparent that a seizure-free outcome may be achieved in some patients (Burgerman et al., 1995; Pacia et al., 1996), suggesting the presence of a truly localized non-lesional temporal lobe epilepsy. The fact that, in two of the present patients, seizures were partly reflex in character (musicogenic and eating-induced) might point to an idiopathic functional aetiology.

Electroclinical syndromes of temporal lobe epilepsy are well described in association with occipital (Salanova et al., 1992; Williamson et al., 1992a) and parietal (Williamson et al., 1992b) lesions, with occasional successful outcomes after anterior temporal lobectomy. In the present study, the two patients with occipital lesions both experienced an initial satisfactory outcome after temporal lobe surgery. Eventual relapse was located electroclinically in the frontocentral and temporal regions but, as resection of the lesions was not considered feasible, it was not possible to define the propagation patterns of the recurrent seizures more accurately with intracranial EEG. Hypothalamic hamartomas may, similarly, have electroclinically defined temporal lobe seizures but the surgical outcome after temporal resection has not been successful (Cascino et al., 1993). It is now evident that intrinsic epileptogenesis with subsequent spread to the cortex is the likely physiology of most seizures in these patients (Munari et al., 1995; Kuzniecky et al., 1997).

The majority of patients with DNT had evidence of focal and multifocal epileptogenesis remote from the resection. Many had multiple seizure types, generalized EEG abnormalities and major cognitive and behavioural disturbance that did not appear to simply reflect the effects of epilepsy from an early age. Previous reports of DNT (Daumas-Duport et al., 1988; Raymond et al., 1994b) highlight this entity as a common and surgically treatable cause of refractory partial epilepsy. The present findings suggest that the clinical correlations of DNT are wider and include conditions amounting, in some cases, to epileptic encephalopathies. Cortical dysplasia frequently occurs as part of the pathological spectrum of DNT (Daumas-Duport et al., 1988; Prayson et al., 1993, 1996; Raymond et al., 1994; Taratuto et al., 1995), and this is a possible explanation for extratemporal epileptogenesis and the other abnormalities in these cases. The association with cortical dysplasia and the known extensive disturbances in neuronal circuitry (Raymond et al., 1994b; Sisodiya et al., 1997) in this setting may provide an explanation for the widespread extratemporal functional abnormalities in these patients. This was confirmed in one patient (Case 38) in whom the pathological demonstration of a band of cortical dysplasia extending beyond the resection into the frontal region correlated with seizure onset on subdural EEG recording. It is possible that a similar mechanism, rather than residual microscopic tumour, could account for persistent seizures arising adjacent to an apparently complete tumour removal as judged by MRI. In fact, subtotal resection of DNT has been associated with a seizure-free outcome in several cases of our series (Kirkpatrick et al., 1993) and others (Daumas-Duport et al., 1988). The concept of `mirror focus' has been invoked as a possible mechanism for persistent seizures after resection of a unilateral lesion. In patients with temporal glial tumours, lesions that are unlikely to occur bilaterally, Morrell reported evidence of contralateral seizure generation in 15% (Morrell, 1985). Failure of contralateral seizures to disappear after surgical resection of the primary epileptogenic region was correlated with the frequency of seizures and the duration of epilepsy. Our results suggest that this mechanism may be less important in patients with DNT. Only one patient (Case 38) had evidence of seizure generation in the contralateral temporal lobe, and this was regarded as non-habitual, occurring in the context of antiepileptic drug reduction during EEG telemetry.

Recurrence of seizures after a seizure-free period of at least 1 year was most frequently encountered in patients with MTS. This finding agrees with that of Berkovic and colleagues (Berkovic et al., 1995) and Spencer (Spencer, 1996), who also reported that late recurrence occurred virtually only in MTS. An important mechanism which has been suggested for late recurrence is maturing epileptogenicity in a surgical scar. However, as most late relapses occurred in MTS patients and not in the others, this suggests that operative trauma is less likely. In fact, patients with DNT had the greatest amount of periresection encephalomalacia but none relapsed after an initially seizure-free interval. Furthermore, in the MTS patients, late relapse from the contralateral hemisphere was disproportionately more frequent than relapse from the ipsilateral hemisphere. In the four MTS patients who had late relapse from the area of previous surgery, one (Case 11) also had seizures arising from the frontal region, remote from the resection. In the other three patients (two with new seizure types), it is possible that a surgical scar could have contributed to seizure relapse. This mechanism is actually suggested by Case 7 (left MTS), in whom haemorrhagic infarction developed adjacent to the operative site (demonstrated by CT on the fifth postoperative day). Although clinical resolution was complete and the reassessment MRI was within normal limits, new postoperative partial seizures consisting of speech arrest could reasonably be attributed to the effects of an operative insult. Apart from these few patients, however, it appears that developing epileptogenesis related to a surgical scar is an unlikely explanation for recurrent seizures arising adjacent to the resection.

Conclusions

The present report outlines the electroclinical patterns of relapse following surgery for temporal lobe epilepsy. The findings highlight the prominence of extrahippocampal and extratemporal epileptogenesis in the majority of surgical failures, irrespective of pathology. Certain features, such as the presence of acoustic auras and ictal EEG onsets suggestive of neocortical seizure origin, could have drawn attention to the possibility of a poor outcome. In many patients, however, operations were performed after electroclinical localization to the mesial temporal region, often, in the case of MTS failures, supported by MRI evidence of unilateral hippocampal atrophy. It is probable that emerging MRI techniques, such as surface coil studies (Barkovich et al., 1995) and curvilinear reformatting (Bastos et al., 1999), may demonstrate subtle cortical abnormalities that may be responsible for postoperative seizures. Similarly, functional neuroimaging with [11C]flumazenil-PET may delineate neuronal deficits suggestive of occult malformation of cortical development in patients with both focal pathology and normal MRI (Richardson et al., 1998). Until such techniques are evaluated in prospective studies of outcome after surgery, patients need to be counselled that, with the available preoperative investigative strategies, relapse may be unpredictable. At present, a normal preoperative high-resolution MRI may be regarded as an indicator for `normal or non-specific pathology' in the majority of patients, and such cases should be approached with caution, as even a clearly defined hippocampal epilepsy on depth recording is not a guarantee of freedom from seizures, the attendant risks of altering seizure propagation patterns and potentially leading to more severe seizures. Finally, with the growing appreciation of the imaging characteristics of DNT and its benign nature, patients with evidence of symptomatic generalized epilepsy should be spared the hazards of a resection which is unlikely to benefit their seizures.

Table 1

Preoperative investigations in patients with MTS at pathology

Case no., age at onset (years), age at surgery (years) History Neuropsychology Neurophysiology Imaging 
AED = anti-epilepsy drug; AEG = air encephalogram; AT = anterior temporal; CPS = complex partial seizure(s); FC = febrile convulsion(s); FDG-PET = [18F]fluorodeoxyglucose-PET; FO = foramen ovale; GTC = generalized tonic–clonic seizure(s); L = left; MesT = mesial temporal; MT = mid-temporal; R = right; SP = simple partial seizure(s); V(P)IQ = verbal (performance) IQ. 
1, 7, 33 Prolonged FC × 2 at 15 and 27 months VIQ 91, PIQ 90. Severe verbal memory deficit. Amytal: normal memory on R, impaired on L Interictal: L AT spikes. FO telemetry: 11 seizures associated with high-frequency MesT spikes CT: dilated L temporal horn 
2, 5, 24 Prolonged FC at 14 months. R facial weakness and ↓ R arm swing VIQ 86, PIQ 97. Normal memory. Amytal: both sides support memory, less on L Interictal: L AT temporal spikes in sleep. FO telemetry: 7 seizures: 6 L MesT onset, 1 bilateral changes 20 s after clinical onset (related to AED withdrawal?) MRI: L hippocampal atrophy 
3, 6, 21 No FC VIQ 93, PIQ 72. Non-verbal memory impaired. Amytal: poor visual memory function on R, memory generally poor Interictal sphenoidal: R fronto- temporal spikes of varying max. amplitude in prefrontal, sphenoidal, MT and sylvian electrodes. FO telemetry: 6 seizures: clear onset R FO bundle but initial changes more prominent in superficial contacts CT normal 
4, 10, 27 No FC VIQ 84, PIQ 74. Normal memory. Amytal: poor memory function on R Interictal sphenoidal: R sphenoidal, MT and sylvian spikes. Scalp telemetry: 3 seizures R anterior quadrant slowing at onset MRI: R hippocampal atrophy 
5, 2, 13 Prolonged FC with R-sided paralysis at 18 months VIQ 82, PIQ 85. Memory impaired but severe attentional difficulty Interictal: left temporal background abnormality with loss of normal rhythms; did not cooperate with seizure recording MRI: severe L hippocampal atrophy + L temporal neocortex appears atrophic 
6, 2½, 33 Afebrile status epilepticus at 30 months (30 min), CPS thereafter VIQ 114, PIQ 103. Verbal memory normal, non-verbal memory impaired. Amytal: both hemispheres support memory but less well on R Interictal: R AT spikes. FO telemetry: 6 CPS: generalized electrodecrement at onset then bilateral FO changes but activity earliest, greatest amplitude R AT scalp electrode MRI: R hippocampal atrophy 
7, 17, 25 Several brief FC 1–4 years old. Brother L MTS, seizure-free 15 years after operation for L MTS. VIQ 91, PIQ 114. Slightly impaired verbal memory. Amytal: both sides support memory adequately Interictal EEG: L AT spikes. MRI: L hippocampal atrophy. FDG-PET: L temporal hypometabolism 
8, 11, 23 No FC. Encephalopathic illness at age 10: headache, drowsy, twitching L limbs × 24 h; CT and CSF normal VIQ 73, PIQ 126. Low VIQ thought to reflect in part low educational achievement. Verbal memory impaired, non-verbal memory severely impaired. Amytal: borderline memory on both sides, right worse than left Interictal: marked background abnormality R temporal region, independent discharges R and L sphenoidal. FO telemetry: 3 seizures, focal onset R MesT CT: dilatation R temporal horn of lateral ventricle 
9, 9, 29 Prolonged FC at 16 months VIQ 82, PIQ 94. Slightly impaired verbal memory. Amytal: adequate memory both sides Interictal: independent AT spikes, L > R. FO telemetry: 4 seizures, all L MesT MRI: L hippocampal atrophy 
10, 7, 18 Generalized TC seizures × 2 at age 7, complicated by mild L hemiparesis VIQ 77, PIQ 81. Impaired non- verbal memory. Amytal: normal L- sided memory, impaired R-sided memory Interictal: R temporal background abnormality; R temporal spikes max. mid-temporal-sylvian. Depth electrodes: 5 seizures: 3 R amygdala, 2 diffuse R temporal subdural and all of R temporal depth electrodes MRI: R hippocampal atrophy + mild diffuse atrophy of R fronto- temporal region 
11, 9, 19 No FC VIQ 82, PIQ 96. Impaired verbal memory Interictal: L AT spikes, occasional L frontal spikes. FO telemetry: 3 CPS, all L MesT MRI: L hippocampal atrophy. FDG-PET: L temporal hypometabolism 
12, 6, 33 Measles, high fever age 6, CPS began during this illness. No suggestion of encephalitis VIQ 87, PIQ 96. Marked non-verbal memory impairment. Amytal: R hemisphere unable to support memory Interictal: R temporal slow activity, single L AT spike and wave discharge. FO telemetry: 11 seizures: 4 auras no change, 3 auras (same) R MesT. 4 CPS: R MesT MRI: R hippocampal atrophy. FDG-PET: R temporal hypometabolism 
13, 9/12, 35 Prolonged FC with R paresis, CPS from that time VIQ 90 PIQ 92. Impaired verbal memory, slight impairment of non- verbal memory: Amytal: borderline memory on both sides Interictal: L AT spikes, scalp telemetry 2 seizures: L hemisphere slowing at onset AEG: L temporal horn dilatation. Xenon CT: L temporal cortical hypoperfusion 
14, 6, 27 Measles aged 6, nocturnal GTC, postictal confusion 3 days. 3 months later CPS begin VIQ 95, PIQ 90. Non-verbal memory moderately impaired, verbal memory mildly impaired. Amytal: both sides support memory adequately Interictal sphenoidal: independent temporal discharges more numerous and extensive on R. FO telemetry: 5 seizures R MesT CT scan and AEG: dilatation R temporal horn 
15, 6, 14 Several brief FC age 6 months to 4 years VIQ 82, PIQ 96. Impaired verbal memory Interictal: almost continuous spike and wave L AT CT scan: dilated left temporal horn 
16, 11/12, 38 Prolonged GTC after measles/pertussis vaccine, CPS thereafter VIQ 95, PIQ 113. Moderate impairment of verbal and non-verbal memory. Amytal: adequate memory on R Interictal: L AT spikes. FO telemetry: 5 seizures, all L MesT CT scan; L hemisphere atrophy 
17, 14, 29 No FC VIQ 105, PIQ 87. Moderate impairment of verbal/non-verbal memory. Amytal: poor memory on R, adequate on L Interictal: R AT spikes. Scalp telemetry: 2 seizures: R anterior quadrant lateralization CT scan normal 
18, 10/12, 21 Several GTC when unwell/fever up to age 4, then CPS, continuing GTC VIQ 75, PIQ 76. Global impairment. Amytal: residual memory R hemisphere Interictal: irregular fast generalized spike-wave. Focal positive spikes mid-parietal. FO telemetry: 16 SP and CPS, left MesT MRI: L anterior hippocampal atrophy FDG-PET: normal 
19, 2, 26 Febrile status epilepticus with R paresis at 12 months VIQ 91, PIQ 83. Mild impairment verbal/non-verbal memory. Amytal: memory adequate both sides Interictal: bifrontal discharges. FO telemetry: 5 seizures: L MesT onset CT scan/AEG dilatation L temporal horn. Interictal SPECT: focal L anterior temporal hypoperfusion 
20, 7, 24 No FC VIQ 79, PIQ 111. Residual memory on R, none on L Interictal sphenoidal: independent bilateral temporosylvian spikes, prominent slow activity R MT- sylvian. FO telemetry: 4 seizures R MesT CT scan normal 
Case no., age at onset (years), age at surgery (years) History Neuropsychology Neurophysiology Imaging 
AED = anti-epilepsy drug; AEG = air encephalogram; AT = anterior temporal; CPS = complex partial seizure(s); FC = febrile convulsion(s); FDG-PET = [18F]fluorodeoxyglucose-PET; FO = foramen ovale; GTC = generalized tonic–clonic seizure(s); L = left; MesT = mesial temporal; MT = mid-temporal; R = right; SP = simple partial seizure(s); V(P)IQ = verbal (performance) IQ. 
1, 7, 33 Prolonged FC × 2 at 15 and 27 months VIQ 91, PIQ 90. Severe verbal memory deficit. Amytal: normal memory on R, impaired on L Interictal: L AT spikes. FO telemetry: 11 seizures associated with high-frequency MesT spikes CT: dilated L temporal horn 
2, 5, 24 Prolonged FC at 14 months. R facial weakness and ↓ R arm swing VIQ 86, PIQ 97. Normal memory. Amytal: both sides support memory, less on L Interictal: L AT temporal spikes in sleep. FO telemetry: 7 seizures: 6 L MesT onset, 1 bilateral changes 20 s after clinical onset (related to AED withdrawal?) MRI: L hippocampal atrophy 
3, 6, 21 No FC VIQ 93, PIQ 72. Non-verbal memory impaired. Amytal: poor visual memory function on R, memory generally poor Interictal sphenoidal: R fronto- temporal spikes of varying max. amplitude in prefrontal, sphenoidal, MT and sylvian electrodes. FO telemetry: 6 seizures: clear onset R FO bundle but initial changes more prominent in superficial contacts CT normal 
4, 10, 27 No FC VIQ 84, PIQ 74. Normal memory. Amytal: poor memory function on R Interictal sphenoidal: R sphenoidal, MT and sylvian spikes. Scalp telemetry: 3 seizures R anterior quadrant slowing at onset MRI: R hippocampal atrophy 
5, 2, 13 Prolonged FC with R-sided paralysis at 18 months VIQ 82, PIQ 85. Memory impaired but severe attentional difficulty Interictal: left temporal background abnormality with loss of normal rhythms; did not cooperate with seizure recording MRI: severe L hippocampal atrophy + L temporal neocortex appears atrophic 
6, 2½, 33 Afebrile status epilepticus at 30 months (30 min), CPS thereafter VIQ 114, PIQ 103. Verbal memory normal, non-verbal memory impaired. Amytal: both hemispheres support memory but less well on R Interictal: R AT spikes. FO telemetry: 6 CPS: generalized electrodecrement at onset then bilateral FO changes but activity earliest, greatest amplitude R AT scalp electrode MRI: R hippocampal atrophy 
7, 17, 25 Several brief FC 1–4 years old. Brother L MTS, seizure-free 15 years after operation for L MTS. VIQ 91, PIQ 114. Slightly impaired verbal memory. Amytal: both sides support memory adequately Interictal EEG: L AT spikes. MRI: L hippocampal atrophy. FDG-PET: L temporal hypometabolism 
8, 11, 23 No FC. Encephalopathic illness at age 10: headache, drowsy, twitching L limbs × 24 h; CT and CSF normal VIQ 73, PIQ 126. Low VIQ thought to reflect in part low educational achievement. Verbal memory impaired, non-verbal memory severely impaired. Amytal: borderline memory on both sides, right worse than left Interictal: marked background abnormality R temporal region, independent discharges R and L sphenoidal. FO telemetry: 3 seizures, focal onset R MesT CT: dilatation R temporal horn of lateral ventricle 
9, 9, 29 Prolonged FC at 16 months VIQ 82, PIQ 94. Slightly impaired verbal memory. Amytal: adequate memory both sides Interictal: independent AT spikes, L > R. FO telemetry: 4 seizures, all L MesT MRI: L hippocampal atrophy 
10, 7, 18 Generalized TC seizures × 2 at age 7, complicated by mild L hemiparesis VIQ 77, PIQ 81. Impaired non- verbal memory. Amytal: normal L- sided memory, impaired R-sided memory Interictal: R temporal background abnormality; R temporal spikes max. mid-temporal-sylvian. Depth electrodes: 5 seizures: 3 R amygdala, 2 diffuse R temporal subdural and all of R temporal depth electrodes MRI: R hippocampal atrophy + mild diffuse atrophy of R fronto- temporal region 
11, 9, 19 No FC VIQ 82, PIQ 96. Impaired verbal memory Interictal: L AT spikes, occasional L frontal spikes. FO telemetry: 3 CPS, all L MesT MRI: L hippocampal atrophy. FDG-PET: L temporal hypometabolism 
12, 6, 33 Measles, high fever age 6, CPS began during this illness. No suggestion of encephalitis VIQ 87, PIQ 96. Marked non-verbal memory impairment. Amytal: R hemisphere unable to support memory Interictal: R temporal slow activity, single L AT spike and wave discharge. FO telemetry: 11 seizures: 4 auras no change, 3 auras (same) R MesT. 4 CPS: R MesT MRI: R hippocampal atrophy. FDG-PET: R temporal hypometabolism 
13, 9/12, 35 Prolonged FC with R paresis, CPS from that time VIQ 90 PIQ 92. Impaired verbal memory, slight impairment of non- verbal memory: Amytal: borderline memory on both sides Interictal: L AT spikes, scalp telemetry 2 seizures: L hemisphere slowing at onset AEG: L temporal horn dilatation. Xenon CT: L temporal cortical hypoperfusion 
14, 6, 27 Measles aged 6, nocturnal GTC, postictal confusion 3 days. 3 months later CPS begin VIQ 95, PIQ 90. Non-verbal memory moderately impaired, verbal memory mildly impaired. Amytal: both sides support memory adequately Interictal sphenoidal: independent temporal discharges more numerous and extensive on R. FO telemetry: 5 seizures R MesT CT scan and AEG: dilatation R temporal horn 
15, 6, 14 Several brief FC age 6 months to 4 years VIQ 82, PIQ 96. Impaired verbal memory Interictal: almost continuous spike and wave L AT CT scan: dilated left temporal horn 
16, 11/12, 38 Prolonged GTC after measles/pertussis vaccine, CPS thereafter VIQ 95, PIQ 113. Moderate impairment of verbal and non-verbal memory. Amytal: adequate memory on R Interictal: L AT spikes. FO telemetry: 5 seizures, all L MesT CT scan; L hemisphere atrophy 
17, 14, 29 No FC VIQ 105, PIQ 87. Moderate impairment of verbal/non-verbal memory. Amytal: poor memory on R, adequate on L Interictal: R AT spikes. Scalp telemetry: 2 seizures: R anterior quadrant lateralization CT scan normal 
18, 10/12, 21 Several GTC when unwell/fever up to age 4, then CPS, continuing GTC VIQ 75, PIQ 76. Global impairment. Amytal: residual memory R hemisphere Interictal: irregular fast generalized spike-wave. Focal positive spikes mid-parietal. FO telemetry: 16 SP and CPS, left MesT MRI: L anterior hippocampal atrophy FDG-PET: normal 
19, 2, 26 Febrile status epilepticus with R paresis at 12 months VIQ 91, PIQ 83. Mild impairment verbal/non-verbal memory. Amytal: memory adequate both sides Interictal: bifrontal discharges. FO telemetry: 5 seizures: L MesT onset CT scan/AEG dilatation L temporal horn. Interictal SPECT: focal L anterior temporal hypoperfusion 
20, 7, 24 No FC VIQ 79, PIQ 111. Residual memory on R, none on L Interictal sphenoidal: independent bilateral temporosylvian spikes, prominent slow activity R MT- sylvian. FO telemetry: 4 seizures R MesT CT scan normal 
Table 2

Preoperative investigations in patients with NSP

Case no., age at onset (years), age at surgery (years) History Neuropsychology Neurophysiology Imaging 
For explanation of abbreviations, see footnote to Table 1
21, 19, 34 – VIQ 92, PIQ 97. Impaired visual memory. Amytal: poor memory R, adequate L Interictal EEG: focal AT, MT and posterior temporal spikes. Depth recording: 4 seizures, all R hippocampal CT normal 
22, 7, 19 – VIQ 82, PIQ 97. Impaired verbal memory. Amytal poor memory L, adequate R Interictal EEG: L AT and MT spikes. FO telemetry: 5 seizures: L temporal onset maximum amplitude of initial discharge at superficial contacts MRI normal 
23, 13, 27 – VIQ 72, PIQ 93. Impaired verbal memory. Amytal: little memory L, normal memory R Interictal EEG: focal AT. FO telemetry: 4 seizures: L temporal onset maximum amplitude of initial discharge at superficial contacts MRI normal 
24, 25, 39 – VIQ 99, PIQ 92. Verbal and visual memory impairment. Amytal: poor memory L, borderline memory R Interictal: L anterior temporal spikes. FO telemetry: 6 seizures, all L MesT MRI normal 
25, 13, 19 – VIQ 87, PIQ 92. Normal memory. Amytal: mixed dominance, both sides support memory adequately R temporal background abnormality. R AT and sylvian spikes. FO: R temporal onset of max. amplitude superficial contacts CT: dilated R temporal horn 
26, 27, 46 – VIQ 84, PIQ 86. Impaired visual memory and impaired perceptual/ constructural tasks: ? additional R centroparietal dysfunction Interictal EEG: R temporal > sylvian > frontal spiking. FO telemetry: 8 seizures: 5 R MesT, 3 R MesT 3 s after clinical onset CT: dilated R temporal horn 
27, 20, 34 Occasional nocturnal GTC for 5 years, then CPS VIQ 97, PIQ 106. Normal memory. Amytal: no memory R Interictal EEG: R AT and MT spikes. Depth recording: 6 seizures: all R hippocampal spreading to R amygdala and then to lateral temporal neocortex CT normal 
28, 10, 23 – VIQ 87, PIQ 90. Impaired visual memory Continuous spiking R AT CT: dilated R temporal horn 
29, 4, 25 – VIQ 86, PIQ 107. Normal memory, Amytal: adequate memory L, little memory R Interictal: R posterior temporal spikes. Depth recording: 5 seizures: R hippocampal onset MRI: lesion located in grey and white matter of R occipital lobe consistent with low-grade tumour 
30, 3, 23 – VIQ 110, PIQ 102. Normal memory FO telemetry: seizure onset maximum L posterior temporal scalp electrode. Depth recording: 4 seizures: all L posterior hippocampal onset MRI: localized atrophy L occipital lobe with dilated occipital horn of lateral ventricle 
31, 19, 40 4 GTC in 2 years, then CPS VIQ 93, PIQ 92. Impaired visual memory Interictal EEG: R MT spikes. Depth recording: R hippocampal onset CT: inferolateral R temporal low-density area 
32, 4, 18 Onset with gelastic seizures VIQ 78, PIQ 92. Severe impairment of verbal memory Interictal EEG: multifocal and generalized discharges, maximum L AT FO telemetry: 6 seizures, all L MesT but 3-4 s after clinical onset MRI: hypothalamic hamartoma 
33, 26, 38 1-2 GTC/year for 7 years, then CPS VIQ 99, PIQ 116. Impaired verbal memory and verbal fluency Interictal EEG: L AT spikes, occasional L frontal spikes. FO telemetry: 5 seizures, all L MesT CT/AEG: dilated L temporal horn 
34, 13, 36 Status epilepticus at 13 with transient R hemiparesis, CPS. thereafter VIQ 87, PIQ 93. Bilateral memory impairment. Amytal: bilateral memory impairment worse on L Interictal EEG: focal L sphenoidal, AT and sylvian spikes. Bilateral frontal sharp and slow activity in sleep. FO telemetry: 6 seizures, all L MesT CT: dilated R temporal horn 
Case no., age at onset (years), age at surgery (years) History Neuropsychology Neurophysiology Imaging 
For explanation of abbreviations, see footnote to Table 1
21, 19, 34 – VIQ 92, PIQ 97. Impaired visual memory. Amytal: poor memory R, adequate L Interictal EEG: focal AT, MT and posterior temporal spikes. Depth recording: 4 seizures, all R hippocampal CT normal 
22, 7, 19 – VIQ 82, PIQ 97. Impaired verbal memory. Amytal poor memory L, adequate R Interictal EEG: L AT and MT spikes. FO telemetry: 5 seizures: L temporal onset maximum amplitude of initial discharge at superficial contacts MRI normal 
23, 13, 27 – VIQ 72, PIQ 93. Impaired verbal memory. Amytal: little memory L, normal memory R Interictal EEG: focal AT. FO telemetry: 4 seizures: L temporal onset maximum amplitude of initial discharge at superficial contacts MRI normal 
24, 25, 39 – VIQ 99, PIQ 92. Verbal and visual memory impairment. Amytal: poor memory L, borderline memory R Interictal: L anterior temporal spikes. FO telemetry: 6 seizures, all L MesT MRI normal 
25, 13, 19 – VIQ 87, PIQ 92. Normal memory. Amytal: mixed dominance, both sides support memory adequately R temporal background abnormality. R AT and sylvian spikes. FO: R temporal onset of max. amplitude superficial contacts CT: dilated R temporal horn 
26, 27, 46 – VIQ 84, PIQ 86. Impaired visual memory and impaired perceptual/ constructural tasks: ? additional R centroparietal dysfunction Interictal EEG: R temporal > sylvian > frontal spiking. FO telemetry: 8 seizures: 5 R MesT, 3 R MesT 3 s after clinical onset CT: dilated R temporal horn 
27, 20, 34 Occasional nocturnal GTC for 5 years, then CPS VIQ 97, PIQ 106. Normal memory. Amytal: no memory R Interictal EEG: R AT and MT spikes. Depth recording: 6 seizures: all R hippocampal spreading to R amygdala and then to lateral temporal neocortex CT normal 
28, 10, 23 – VIQ 87, PIQ 90. Impaired visual memory Continuous spiking R AT CT: dilated R temporal horn 
29, 4, 25 – VIQ 86, PIQ 107. Normal memory, Amytal: adequate memory L, little memory R Interictal: R posterior temporal spikes. Depth recording: 5 seizures: R hippocampal onset MRI: lesion located in grey and white matter of R occipital lobe consistent with low-grade tumour 
30, 3, 23 – VIQ 110, PIQ 102. Normal memory FO telemetry: seizure onset maximum L posterior temporal scalp electrode. Depth recording: 4 seizures: all L posterior hippocampal onset MRI: localized atrophy L occipital lobe with dilated occipital horn of lateral ventricle 
31, 19, 40 4 GTC in 2 years, then CPS VIQ 93, PIQ 92. Impaired visual memory Interictal EEG: R MT spikes. Depth recording: R hippocampal onset CT: inferolateral R temporal low-density area 
32, 4, 18 Onset with gelastic seizures VIQ 78, PIQ 92. Severe impairment of verbal memory Interictal EEG: multifocal and generalized discharges, maximum L AT FO telemetry: 6 seizures, all L MesT but 3-4 s after clinical onset MRI: hypothalamic hamartoma 
33, 26, 38 1-2 GTC/year for 7 years, then CPS VIQ 99, PIQ 116. Impaired verbal memory and verbal fluency Interictal EEG: L AT spikes, occasional L frontal spikes. FO telemetry: 5 seizures, all L MesT CT/AEG: dilated L temporal horn 
34, 13, 36 Status epilepticus at 13 with transient R hemiparesis, CPS. thereafter VIQ 87, PIQ 93. Bilateral memory impairment. Amytal: bilateral memory impairment worse on L Interictal EEG: focal L sphenoidal, AT and sylvian spikes. Bilateral frontal sharp and slow activity in sleep. FO telemetry: 6 seizures, all L MesT CT: dilated R temporal horn 
Table 3

Persistent/recurrent seizures: preoperative details and postoperative clinical, imaging and neurophysiological investigations

Case no., side of surgery, pathology Preoperative clinical details Postoperative clinical details EEG MRI Electroclinical classification 
AH = amygdalo-hippocampectomy; BG = background; LOC = loss of consciousness; NES = non-epileptic seizure; NFV = non-forced version; NS = non-specific. For explanation of other abbreviations, see footnote to Table 1
1, L, MTS CPS: scrotal sensation, fear, LOC, postictal dysphasia. No GTC 7 years seizure-free, CPS: aura, L limb automatisms, R arm dystonia, version to R before generalization Ictal: spike–wave L temporo- sylvian region 20 s after clinical onset L hippocampus completely resected. R mesial temporal structures normal Ipsilateral temporal 
2, L, MTS CPS: epigastric aura, early NFV to left, then version to R before secondary generalization. Postictal dysphasia CPS: non-specific aura, NFV to left, R arm dystonia, then version to R and generalization Ictal: rhythmic theta left mid- temporal and sylvian region 20 s after clinical onset L hippocampus completely resected. R temporal structures normal Ipsilateral temporal 
3, R, MTS CPS: no aura, stares, speech arrest, generalization CPS: buzzing in ears, dystonia L limbs, generalization Ictal: rhythmic spikes R mid- temporal and sylvian region 15 s before clinical onset R hippocampus completely resected. L mesial temporal structures normal Ipsilateral temporal 
4, R, MTS CPS: buzzing in ears, hums, head turns to L, posturing of L arm, generalizes CPS: buzzing in ears, abrupt version of head to L and tonic posturing of L arm Ictal: generalized decrement at clinical onset followed by low-voltage fast activity arising over posterior contacts of R temporosylvian subdural mat. Depth electrode in residual hippocampus not involved Residual R hippocampus, increased T2 signal, no neocortical lesion identified Ipsilateral temporal 
5, L, MTS CPS: no aura, stares, dystonia R arm. Special schooling, IQ <70 Seizure-free 14 months. CPS: no warning, dystonia R limbs, generalization, Todd's paresis R limbs Interictal (preop.) left temporal BG abnormality, L anterior/ mid-temporal spikes; Interictal (postop.) L midtemporal spikes Preop MRI: L hippocampal atrophy, mild atrophy entire L temporal lobe. Postop. MRI: complete L hippocampal resection Ipsilateral temporal 
6, R, MTS CPS: epigastric aura, stares, automatisms Seizure-free 8 months. SP: ringing noise in head. Nocturnal secondary generalized seizures, 4/month Interictal (postop.) runs of R centrosylvian sharp/slow complexes Complete R hippocampal resection; no other abnormality Ipsilateral temporal 
7, L, MTS CPS: no aura, stares, jargon dysphasia. L hemisphere dominant for speech on Wada test Seizure-free 8 months. Partial seizures: speech arrest. Generalized with postictal dysphasia Ictal (preop.) FO electrodes: left mesial temporal onset. Interictal (postop.) L mid- temporal spike–wave Postop. CT (day 3) haematoma adjacent to resection. Preop. MRI: L hippocampal atrophy. Postop. MRI: L hippocampus completely resected. CT abnormality resolved Ipsilateral temporal 
8, R, MTS (AH) CPS: epigastric aura, mutters, generalization Seizure-free 2 years. Then CPS: no warning, automatisms, L arm dystonia, generalization Ictal: R anterior temporal spikes at clinical onset Residual R hippocampus ++, increased T2 signal Ipsilateral temporal 
9, L, MTS (AH) CPS: epigastric aura, NFV to L, automatisms L arm, postictal dysphasia Three seizures in 5 years. Then CPS, often nocturnal. Epigastric aura, limb automatisms, postictal dysphasia Ictal: rhythmic spikes confined to L midtemporal-sylvian region for initial 40 s Complete L hippocampal resection, no other abnormality Ipsilateral temporal 
10, R, MTS CPS: olfactory aura, staring, dystonia L arm, version to L and generalization Olfactory aura and staring attacks gone. Generalized seizures preceded by jerking L arm Ictal (preop.) depth/subdural recording: 3 seizures R amygdala, 2 regional onset mid/posterior temporal subdural. Interictal: R midtemporal/sylvian spikes Preop. MRI: R hippocampal atrophy, mild atrophy R frontotemporal region. Postop. MRI: complete R hippocampal resection Ipsilateral hemisphere 
11, L, MTS CPS: epigastric aura, dystonia R arm, version of head to L before generalization. Postictal dysphasia Seizure-free 1 year. CPS: (i) nocturnal, panic, jerking of limbs, tonic contraction R face; (ii) buzzing sensation R ear, sensation R body and contraction of R limbs, version to R and generalization Ictal: (i) Rhythmic sharp and slow activity L frontal region 1 min before clinical onset; (ii) rhythmic spikes left sylvian at clinical onset. Complete L hippocampal resection. No other abnormality Ipsilateral frontal, ipsilateral temporal 
12, R, MTS CPS: non-specific aura, stares, NFV to R, limb stiffening Episodes of complex partial status (confusional state), nocturnal GTC 2/month, (not visualized), preop. CPS gone Ictal: rhythmic R frontal delta during CP status. Run of sharp waves over R frontal region, thought to represent subclinical seizure Small amount of residual hippo- campus on R, L mesial temporal structures normal. No frontal abnormality demonstrated Ipsilateral frontal 
13, L, MTS (AH) CPS: epigastric aura, stare, generalization begins in R limbs, postictal dysphasia, usually daytime Seizure-free 5 months, nocturnal seizures, NFV to R, secondary generalization Ictal: rhythmic spikes R anterior temporal region L hippocampus complete, resection R hippocampus returns high signal on FLAIR sequence Contralateral temporal 
14, R, MTS (AH) CPS: non-specific aura, automatisms, ictal speech, generalization CPS: non-specific aura, stare, loss of awareness Ictal: L mesial temporal onset (FO electrodes) before clinical onset Not done (died during period of reassessment) Contralateral temporal 
15, L, MTS CPS: epigastric aura, L hand automatisms, R arm dystonia, always generalizes Seizure-free apart from occasional auras for 4 years. Then CPS: no aura, NFV to R, automatism of R arm, dystonia L arm, version to L before generalizing Ictal: spike and wave over R mid and anterior temporal region 30 s before clinical onset Complete L hippocampal resection, R hippocampal structures normal (FLAIR) Contralateral temporal 
16, L, MTS CPS: epigastric aura, stares, dystonia of R arm, turns to R before generalizing Seizure-free for 5 years. CPS: no warning, stares, automatisms R hand, NFV to R and late version to left before generalizing Ictal: rhythmic theta R anterior temporal region at clinical onset Almost complete hippocampal resection on L, increased T2 signal R hippocampus Contralateral temporal 
17, R, MTS CPS: epigastric aura, LOC, speech. No GTC CPS: usually from sleep, stares, no aura, dystonia R arm Ictal: rhythmic spikes L anterior temporal region at onset Complete R hippocampal, resection. Small L hippocampus with increased T2 signal Contralateral temporal 
18, L, MTS CPS: epigastric aura, stares. 1-2 GTC/month Nocturnal GTC, 7/week, preop. CPS gone Ictal: generalized polyspike and wave at onset, no focal features Left anterior TL, most of parahippocampal gyrus remaining. No neocortical lesion. R mesial temporal structures normal Generalized 
19, L, MTS (AH) CPS: cephalic aura, stare, dystonia R arm, jargon dysphasia at end. No GTC Seizure-free for 2 years. CPS: (i) no aura, R arm dystonia, LOC, grimace; (ii) initial bipedal cycling movements, L arm dystonia, speech Ictal: (i) L subtemporal strip onset (1 seizure); (ii) simultaneous onset R anterior scalp and R subtemporal strip electrodes (3 seizures) Residual hippocampus on L. Increased T2 signal, R hippocampus normal. No neocortical abnormality in frontal region (FLAIR sequence) Bilateral independent (ipsilateral mesial temporal and contralateral frontal) 
20, R, MTS CPS: noise in head, deaf, epigastric aura, dystonia L arm, late automatism R arm, NFV to R, turns to L before generalizing CPS: (i) noise in head, stares, dystonia L arm, automatisms of R arm; (ii) restless, grunts, flexion of both upper limbs, out of contact with surroundings Ictal: (i) spike and wave R temporal region at clinical onset; (ii) rhythmic sharp activity L anterior temporal region 15 s after clinical onset Much of R hippocampus and parahippocampal gyrus remaining and atrophic, L mesial temporal structures appear normal Bilateral independent temporal 
21, R, NS CPS: no aura, stares, NFV to R, dystonia L arm CPS: stares, slow version of head to left, flexion of arms, facial grimacing and clonic jerking of L arm Ictal: rhythmic spikes R sylvian and mid-temporal region at clinical onset Small residual portion R hippocampal tail, L mesial temporal structures normal Ipsilateral temporal 
22, L, NS CPS: piercing noise in head, stares, oral automatisms, dystonia R arm. Postictal dysphasia CPS piercing noise, numbness of tongue and palate, blinking, clonic jerking R face, numbness R arm and trunk. Postictal R hemiparesis and dysphasia. 70% of attacks associated with eating Ictal: rhythmic discharge R sylvian-midtemporal region at clinical onset Complete L hippocampal resection. No other abnormality Ipsilateral temporal 
23, L, NS CPS: speech arrest, stares, postictal dysphasia. R facial weakness. L hemisphere language dominant Seizure-free 15 months. Partial seizures: episodes of speech arrest, nocturnal generalized seizures Interictal (preop.): diffuse spikes L frontotemporal. Ictal (preop.): ictal onset lateral temporal contacts of FO bundle. Interictal (postop.): L temporosylvian spikes Postop. MRI: L posterior hippo- campal remnant. Remaining L temporal neocortex appears atrophic Ipsilateral temporal 
24, L, NS (AH) CPS: no aura, NFV to left, automatisms of hands CPS: No aura, stares, L arm automatisms, R arm immobility Ictal: rhythmic spikes over left temporal region before clinical onset, remains con- fined to this region for 50 s Residual hippocampus and parahippocampal gyrus on L, no signal change. R temporal structures normal Ipsilateral temporal 
25, R, NS CPS: buzzing noise in head, epigastric sensation, ictal speech. Special school, IQ 80 No change. Erratic aggressive behaviour, interictal psychotic episodes Interictal (preop.): anterior temporal spikes, independent R superior frontal, sylvian and parietal spikes. Ictal (preop.): onset R mid and posterior temporal scalp electrodes. Interictal (postop.): R-sided spikes in same distribution Preop. CT: low-density area medial R temporal lobe (abnormality not confirmed at operation/pathology) Postop. MRI: complete R hippo- campal resection, no other abnormality Ipsilateral hemisphere 
26, R, NS CPS: buzzing in ears, sounds seem louder, can have speech arrest, bilateral reactive automatisms Frequency of seizures reduced from 9/week to 2/month. CPS: buzzing in ears, frequently no warning, stares, dystonia of L arm Ictal: bifrontal delta during CPS, no clear lateralizing changes at onset R hippocampus completely resected. High signal on PD and T2 sequences extending from R basal ganglia to above right lateral ventricle, consistent with infarction Uncertain 
27, R, NS CPS: epigastric aura, stares, automatisms, agitation, frequent micturition Initially 3 generalized nocturnal seizures in first 2 years, then: CPS: loss of awareness, leans to right and flexes both arms in front of face. All nocturnal Ictal: rhythmic theta R superior and mid-frontal electrodes 20 s before seizure onset R hippocampus completely resected. No other abnormality Ipsilateral frontal 
28, R, NS CPS: 80% attacks precipitated by music, stares, speech, automatisms of hands Preop. seizures replaced ascending sensory seizures affecting L arm, with some impairment of awareness Ictal: subdural mat: focal ictal transformation R sensory cortex (hand area, confirmed by electrical stimulation) Residual hippocampus on R, no signal change, no R suprasylvian abnormality Ipsilateral centrosylvian 
29, R, NS CPS: fear, stares, automatisms, NFV to R, dystonia L arm Infrequent nocturnal generalized seizures for 6 years. Then (i) CPS returned, brief, fear; (ii) major fits increased in frequency: stares,version to left, kicking of the legs, dystonic posturing of L arm (all from sleep) Ictal: (i) not telemetered; (ii) rhythmic spike and wave maximum over the R frontal and central regions within 10 s of clinical onset R parieto-occipital lesion, increased signal on PD and T2-weighted images. Grey matter abnormality R occipital lobe. Probable low- grade tumour such as DNT, with associated cortical dysplasia Ipsilateral occipital 
30, L, NS CPS: yellow lights R visual field, epigastric aura, automatisms of L hand, generalization Seizure-free 2 years apart from déjà vu auras. Then CPS: stares, version to L and generalizes. No visual symptoms Interictal (preop) L anterior temporal spikes. Ictal (preop.): subdural and depth: L posterior hippocampal onset. Interictal (postop.): frequent R mid- and posterior temporal discharges Pre- and postop. MRI: dilatation of L occipital horn with localized focal occipital cortical atrophy Ipsilateral occipital 
31, R, NS CPS: No aura, stares, rubs hands,mutters, wanders about CPS: no aura, confusion, automatisms, agitated Ictal: spike and wave L temporal region 4 s after clinical onset R hippocampus completely resected, L temporal structures normal Contralateral temporal 
32, L, NS Gelastic seizures. CPS: non- specific aura, stares. Sometimes accompanied by an attack of laughter New attacks: brief sensation R leg, R face distorts, generalization. Outbursts of inappropriate, hysterical laughter. Generalized seizures Interictal (pre- and postop.) multifocal, generalized discharges, of greatest amplitude L hemisphere. Ictal (preop.): L FO onset just after clinical onset Pre- and postop. MRI: small hypothalamic hamartoma (deeply situated in gland) Multifocal 
33, L, NS CPS: déjà vu, automatisms, version to the right before generalization NES: intermittent shaking of legs, sometimes R arm, fluctuates for up to 1 h, distractible Normal EEG throughout Complete hippocampal resection on L. Normal R temporal structures NES 
34, L, NS (AH) CPS: cephalic aura, reactive automatisms, contortion of R face NES: rhythmic flapping movements of the arms, can be stopped by holding on to the hands Normal EEG throughout Postop. MRI not performed NES 
Case no., side of surgery, pathology Preoperative clinical details Postoperative clinical details EEG MRI Electroclinical classification 
AH = amygdalo-hippocampectomy; BG = background; LOC = loss of consciousness; NES = non-epileptic seizure; NFV = non-forced version; NS = non-specific. For explanation of other abbreviations, see footnote to Table 1
1, L, MTS CPS: scrotal sensation, fear, LOC, postictal dysphasia. No GTC 7 years seizure-free, CPS: aura, L limb automatisms, R arm dystonia, version to R before generalization Ictal: spike–wave L temporo- sylvian region 20 s after clinical onset L hippocampus completely resected. R mesial temporal structures normal Ipsilateral temporal 
2, L, MTS CPS: epigastric aura, early NFV to left, then version to R before secondary generalization. Postictal dysphasia CPS: non-specific aura, NFV to left, R arm dystonia, then version to R and generalization Ictal: rhythmic theta left mid- temporal and sylvian region 20 s after clinical onset L hippocampus completely resected. R temporal structures normal Ipsilateral temporal 
3, R, MTS CPS: no aura, stares, speech arrest, generalization CPS: buzzing in ears, dystonia L limbs, generalization Ictal: rhythmic spikes R mid- temporal and sylvian region 15 s before clinical onset R hippocampus completely resected. L mesial temporal structures normal Ipsilateral temporal 
4, R, MTS CPS: buzzing in ears, hums, head turns to L, posturing of L arm, generalizes CPS: buzzing in ears, abrupt version of head to L and tonic posturing of L arm Ictal: generalized decrement at clinical onset followed by low-voltage fast activity arising over posterior contacts of R temporosylvian subdural mat. Depth electrode in residual hippocampus not involved Residual R hippocampus, increased T2 signal, no neocortical lesion identified Ipsilateral temporal 
5, L, MTS CPS: no aura, stares, dystonia R arm. Special schooling, IQ <70 Seizure-free 14 months. CPS: no warning, dystonia R limbs, generalization, Todd's paresis R limbs Interictal (preop.) left temporal BG abnormality, L anterior/ mid-temporal spikes; Interictal (postop.) L midtemporal spikes Preop MRI: L hippocampal atrophy, mild atrophy entire L temporal lobe. Postop. MRI: complete L hippocampal resection Ipsilateral temporal 
6, R, MTS CPS: epigastric aura, stares, automatisms Seizure-free 8 months. SP: ringing noise in head. Nocturnal secondary generalized seizures, 4/month Interictal (postop.) runs of R centrosylvian sharp/slow complexes Complete R hippocampal resection; no other abnormality Ipsilateral temporal 
7, L, MTS CPS: no aura, stares, jargon dysphasia. L hemisphere dominant for speech on Wada test Seizure-free 8 months. Partial seizures: speech arrest. Generalized with postictal dysphasia Ictal (preop.) FO electrodes: left mesial temporal onset. Interictal (postop.) L mid- temporal spike–wave Postop. CT (day 3) haematoma adjacent to resection. Preop. MRI: L hippocampal atrophy. Postop. MRI: L hippocampus completely resected. CT abnormality resolved Ipsilateral temporal 
8, R, MTS (AH) CPS: epigastric aura, mutters, generalization Seizure-free 2 years. Then CPS: no warning, automatisms, L arm dystonia, generalization Ictal: R anterior temporal spikes at clinical onset Residual R hippocampus ++, increased T2 signal Ipsilateral temporal 
9, L, MTS (AH) CPS: epigastric aura, NFV to L, automatisms L arm, postictal dysphasia Three seizures in 5 years. Then CPS, often nocturnal. Epigastric aura, limb automatisms, postictal dysphasia Ictal: rhythmic spikes confined to L midtemporal-sylvian region for initial 40 s Complete L hippocampal resection, no other abnormality Ipsilateral temporal 
10, R, MTS CPS: olfactory aura, staring, dystonia L arm, version to L and generalization Olfactory aura and staring attacks gone. Generalized seizures preceded by jerking L arm Ictal (preop.) depth/subdural recording: 3 seizures R amygdala, 2 regional onset mid/posterior temporal subdural. Interictal: R midtemporal/sylvian spikes Preop. MRI: R hippocampal atrophy, mild atrophy R frontotemporal region. Postop. MRI: complete R hippocampal resection Ipsilateral hemisphere 
11, L, MTS CPS: epigastric aura, dystonia R arm, version of head to L before generalization. Postictal dysphasia Seizure-free 1 year. CPS: (i) nocturnal, panic, jerking of limbs, tonic contraction R face; (ii) buzzing sensation R ear, sensation R body and contraction of R limbs, version to R and generalization Ictal: (i) Rhythmic sharp and slow activity L frontal region 1 min before clinical onset; (ii) rhythmic spikes left sylvian at clinical onset. Complete L hippocampal resection. No other abnormality Ipsilateral frontal, ipsilateral temporal 
12, R, MTS CPS: non-specific aura, stares, NFV to R, limb stiffening Episodes of complex partial status (confusional state), nocturnal GTC 2/month, (not visualized), preop. CPS gone Ictal: rhythmic R frontal delta during CP status. Run of sharp waves over R frontal region, thought to represent subclinical seizure Small amount of residual hippo- campus on R, L mesial temporal structures normal. No frontal abnormality demonstrated Ipsilateral frontal 
13, L, MTS (AH) CPS: epigastric aura, stare, generalization begins in R limbs, postictal dysphasia, usually daytime Seizure-free 5 months, nocturnal seizures, NFV to R, secondary generalization Ictal: rhythmic spikes R anterior temporal region L hippocampus complete, resection R hippocampus returns high signal on FLAIR sequence Contralateral temporal 
14, R, MTS (AH) CPS: non-specific aura, automatisms, ictal speech, generalization CPS: non-specific aura, stare, loss of awareness Ictal: L mesial temporal onset (FO electrodes) before clinical onset Not done (died during period of reassessment) Contralateral temporal 
15, L, MTS CPS: epigastric aura, L hand automatisms, R arm dystonia, always generalizes Seizure-free apart from occasional auras for 4 years. Then CPS: no aura, NFV to R, automatism of R arm, dystonia L arm, version to L before generalizing Ictal: spike and wave over R mid and anterior temporal region 30 s before clinical onset Complete L hippocampal resection, R hippocampal structures normal (FLAIR) Contralateral temporal 
16, L, MTS CPS: epigastric aura, stares, dystonia of R arm, turns to R before generalizing Seizure-free for 5 years. CPS: no warning, stares, automatisms R hand, NFV to R and late version to left before generalizing Ictal: rhythmic theta R anterior temporal region at clinical onset Almost complete hippocampal resection on L, increased T2 signal R hippocampus Contralateral temporal 
17, R, MTS CPS: epigastric aura, LOC, speech. No GTC CPS: usually from sleep, stares, no aura, dystonia R arm Ictal: rhythmic spikes L anterior temporal region at onset Complete R hippocampal, resection. Small L hippocampus with increased T2 signal Contralateral temporal 
18, L, MTS CPS: epigastric aura, stares. 1-2 GTC/month Nocturnal GTC, 7/week, preop. CPS gone Ictal: generalized polyspike and wave at onset, no focal features Left anterior TL, most of parahippocampal gyrus remaining. No neocortical lesion. R mesial temporal structures normal Generalized 
19, L, MTS (AH) CPS: cephalic aura, stare, dystonia R arm, jargon dysphasia at end. No GTC Seizure-free for 2 years. CPS: (i) no aura, R arm dystonia, LOC, grimace; (ii) initial bipedal cycling movements, L arm dystonia, speech Ictal: (i) L subtemporal strip onset (1 seizure); (ii) simultaneous onset R anterior scalp and R subtemporal strip electrodes (3 seizures) Residual hippocampus on L. Increased T2 signal, R hippocampus normal. No neocortical abnormality in frontal region (FLAIR sequence) Bilateral independent (ipsilateral mesial temporal and contralateral frontal) 
20, R, MTS CPS: noise in head, deaf, epigastric aura, dystonia L arm, late automatism R arm, NFV to R, turns to L before generalizing CPS: (i) noise in head, stares, dystonia L arm, automatisms of R arm; (ii) restless, grunts, flexion of both upper limbs, out of contact with surroundings Ictal: (i) spike and wave R temporal region at clinical onset; (ii) rhythmic sharp activity L anterior temporal region 15 s after clinical onset Much of R hippocampus and parahippocampal gyrus remaining and atrophic, L mesial temporal structures appear normal Bilateral independent temporal 
21, R, NS CPS: no aura, stares, NFV to R, dystonia L arm CPS: stares, slow version of head to left, flexion of arms, facial grimacing and clonic jerking of L arm Ictal: rhythmic spikes R sylvian and mid-temporal region at clinical onset Small residual portion R hippocampal tail, L mesial temporal structures normal Ipsilateral temporal 
22, L, NS CPS: piercing noise in head, stares, oral automatisms, dystonia R arm. Postictal dysphasia CPS piercing noise, numbness of tongue and palate, blinking, clonic jerking R face, numbness R arm and trunk. Postictal R hemiparesis and dysphasia. 70% of attacks associated with eating Ictal: rhythmic discharge R sylvian-midtemporal region at clinical onset Complete L hippocampal resection. No other abnormality Ipsilateral temporal 
23, L, NS CPS: speech arrest, stares, postictal dysphasia. R facial weakness. L hemisphere language dominant Seizure-free 15 months. Partial seizures: episodes of speech arrest, nocturnal generalized seizures Interictal (preop.): diffuse spikes L frontotemporal. Ictal (preop.): ictal onset lateral temporal contacts of FO bundle. Interictal (postop.): L temporosylvian spikes Postop. MRI: L posterior hippo- campal remnant. Remaining L temporal neocortex appears atrophic Ipsilateral temporal 
24, L, NS (AH) CPS: no aura, NFV to left, automatisms of hands CPS: No aura, stares, L arm automatisms, R arm immobility Ictal: rhythmic spikes over left temporal region before clinical onset, remains con- fined to this region for 50 s Residual hippocampus and parahippocampal gyrus on L, no signal change. R temporal structures normal Ipsilateral temporal 
25, R, NS CPS: buzzing noise in head, epigastric sensation, ictal speech. Special school, IQ 80 No change. Erratic aggressive behaviour, interictal psychotic episodes Interictal (preop.): anterior temporal spikes, independent R superior frontal, sylvian and parietal spikes. Ictal (preop.): onset R mid and posterior temporal scalp electrodes. Interictal (postop.): R-sided spikes in same distribution Preop. CT: low-density area medial R temporal lobe (abnormality not confirmed at operation/pathology) Postop. MRI: complete R hippo- campal resection, no other abnormality Ipsilateral hemisphere 
26, R, NS CPS: buzzing in ears, sounds seem louder, can have speech arrest, bilateral reactive automatisms Frequency of seizures reduced from 9/week to 2/month. CPS: buzzing in ears, frequently no warning, stares, dystonia of L arm Ictal: bifrontal delta during CPS, no clear lateralizing changes at onset R hippocampus completely resected. High signal on PD and T2 sequences extending from R basal ganglia to above right lateral ventricle, consistent with infarction Uncertain 
27, R, NS CPS: epigastric aura, stares, automatisms, agitation, frequent micturition Initially 3 generalized nocturnal seizures in first 2 years, then: CPS: loss of awareness, leans to right and flexes both arms in front of face. All nocturnal Ictal: rhythmic theta R superior and mid-frontal electrodes 20 s before seizure onset R hippocampus completely resected. No other abnormality Ipsilateral frontal 
28, R, NS CPS: 80% attacks precipitated by music, stares, speech, automatisms of hands Preop. seizures replaced ascending sensory seizures affecting L arm, with some impairment of awareness Ictal: subdural mat: focal ictal transformation R sensory cortex (hand area, confirmed by electrical stimulation) Residual hippocampus on R, no signal change, no R suprasylvian abnormality Ipsilateral centrosylvian 
29, R, NS CPS: fear, stares, automatisms, NFV to R, dystonia L arm Infrequent nocturnal generalized seizures for 6 years. Then (i) CPS returned, brief, fear; (ii) major fits increased in frequency: stares,version to left, kicking of the legs, dystonic posturing of L arm (all from sleep) Ictal: (i) not telemetered; (ii) rhythmic spike and wave maximum over the R frontal and central regions within 10 s of clinical onset R parieto-occipital lesion, increased signal on PD and T2-weighted images. Grey matter abnormality R occipital lobe. Probable low- grade tumour such as DNT, with associated cortical dysplasia Ipsilateral occipital 
30, L, NS CPS: yellow lights R visual field, epigastric aura, automatisms of L hand, generalization Seizure-free 2 years apart from déjà vu auras. Then CPS: stares, version to L and generalizes. No visual symptoms Interictal (preop) L anterior temporal spikes. Ictal (preop.): subdural and depth: L posterior hippocampal onset. Interictal (postop.): frequent R mid- and posterior temporal discharges Pre- and postop. MRI: dilatation of L occipital horn with localized focal occipital cortical atrophy Ipsilateral occipital 
31, R, NS CPS: No aura, stares, rubs hands,mutters, wanders about CPS: no aura, confusion, automatisms, agitated Ictal: spike and wave L temporal region 4 s after clinical onset R hippocampus completely resected, L temporal structures normal Contralateral temporal 
32, L, NS Gelastic seizures. CPS: non- specific aura, stares. Sometimes accompanied by an attack of laughter New attacks: brief sensation R leg, R face distorts, generalization. Outbursts of inappropriate, hysterical laughter. Generalized seizures Interictal (pre- and postop.) multifocal, generalized discharges, of greatest amplitude L hemisphere. Ictal (preop.): L FO onset just after clinical onset Pre- and postop. MRI: small hypothalamic hamartoma (deeply situated in gland) Multifocal 
33, L, NS CPS: déjà vu, automatisms, version to the right before generalization NES: intermittent shaking of legs, sometimes R arm, fluctuates for up to 1 h, distractible Normal EEG throughout Complete hippocampal resection on L. Normal R temporal structures NES 
34, L, NS (AH) CPS: cephalic aura, reactive automatisms, contortion of R face NES: rhythmic flapping movements of the arms, can be stopped by holding on to the hands Normal EEG throughout Postop. MRI not performed NES 
Table 4

Persistent seizures after temporal resection of DNTs: pre- and postoperative clinical and investigative features

Case no., side of surgery Preoperative clinical Preoperative imaging Preoperative EEG Postoperative clinical Postoperative imaging Postoperative EEG Electroclinical classification 
FSIQ = full scale IQ; LD = learning disability; NES = non-epileptic seizure. For other abbreviations see footnote to Table 1
35, R CPS: stares, automatisms, version to L. VIQ 96, PIQ 69. Verbal and visuospatial memory impaired CT R posterior temporal lesion involving hippocampus posteriorly Interictal: gen. spike and wave > on R, R post- temporal slow, R post- temporal spike and wave in sleep. Ictal (FO): rhythmic discharge R FO electrode (i) Frequent NES: hyperventilation and flailing of arms. (ii) Less frequent CPS: stares, rapid version to L +/– generalization Complete resection back beyond the fornix and to level of posterior horn of R lateral ventricle. No residual tumour Inter-ictal: R post- temporal spikes. Ictal: R frontocentral rhythmic theta 10 s after clinical onset Ipsilateral posterior temporal NES 
36, L CPS: stares, arms flex, head to R, posturing R arm. Sec. GTC seizures. VIQ 93, PIQ 80 CT normal Variable spiking L ant., mid and posterior temporal. FO telemetry: 3/5 seizures L FO onset, 2 similar attacks unclear onset Frequent, mainly nocturnal, seizures: stares, rapid version of head to R and elevation of R arm before generalization MRI: extensive L temporal lobectomy with posterior temporal mass lesion. Mild diffuse L hemisphere atrophy. Signal change L temporo-occipital region Ictal: rhythmic theta L central, spreading to L frontal region. Subdural mat recording: left infero- posterior temporal onset Ipsilateral posterior temporal 
37, L CPS (i) nocturnal jerking R limbs, version of head to R; (ii) fear, aphasia, upper limb automatisms. FSIQ 83, VIQ 74, PIQ 104 CT/angiography: L anterior temporal avascular lesion Diffuse slow. Gen. polyspike and wave. High-amplitude spikes L posterior temporal Frequent nocturnal GTC preceded by version of head to R and jerking of R limbs MRI: extensive L temporal resection, no residual tumour. Gliotic changes posterior to resection margin. Diffuse L hemisphere Frequent spike-wave discharges L mid and posterior temporal electrodes. Gen. symmetrical polyspike and wave atrophy Ipsilateral posterior temporal 
38, R CPS: pale, swallows, talks nonsense, runs. Hyperactive, disruptive, aggressive. Normal intelligence MRI signal abnormality R mesial temporal lobe extending to trigone of R lateral ventricle Slow waves and frequent spike and wave R anterior temporal region (i) Frequent nocturnal seizures: body twists to R, head to L with extension of L arm and flexion of R arm; (ii) confusion, runs MRI: complete resection of R temporal lesion. Simplification of R frontal gyral pattern. L mesial temporal structures normal Ictal: (i) onset anterior contacts of temporo- central mat electrode and proximal contacts of R frontal strip electrode; (ii) rhythmic spikes left anterior temporal Bilateral independent (ipsilateral frontal and contralateral temporal 
39, L Seizures: (i) vacant spells; (ii) drop attacks; (iii) dystonic jerking L limbs; (iv) GTC. FSIQ 60, VIQ 41, PIQ 68. Behavioural problems, possible psychosis CT: L mesial temporal calcified lesion extending to basal ganglia and cerebral peduncle temporal + sylvian spikes Bilateral, independent spikes, generalized slowing + gross L temporal slowing with high-amplitude L Some reduction in seizure frequency, multiple seizure types and psychological problems persist MRI: extensive L temporal lobectomy; no signs of residual tumour at resection margins. Residual tumour deeply at level of basal ganglia L temporal slowing. L suprasylvian spike and wave. Independent R-sided discharges. General discharges in sleep Generalized/ multifocal 
40, L CPS: pale, loss of awareness. GTC. Normal development to age 5, then special school. Age 5, FSIQ 78; age 14, VIQ 45, PIQ 58 CT: L mesial temporal mass extending deep to insula and to basal ganglia Diffuse slow. Multifocal and generalized spikes. Repeated discharges L posterior temporal and parietal Frequent nocturnal generalized seizures with no focal features MRI: residual tumour around margin of left temporal lobectomy, esp. deeply in the L hemisphere and the upper brainstem and thalamus Diffuse background abnormality; multifocal and general spike and wave. Focal L parasagittal discharges Generalized/ multifocal 
41, R Drop attacks. CPS: stares, R facial distortion. GTC. Normal birth, developmental regression from 4 years. Severe LD, IQ < 30 CT: calcified lesion mesial R temporal lobe R and L hemisphere Interictal EEG: bursts of general polyspike and wave, independent Severe cognitive d/sc. Carotid Amytal: suggested independence of L-sided discharges No change; continues with frequent GTC, CPS and drop attacks. hemisphere atrophy. impairment MRI: complete removal of R temporal tumour, probable diffuse R frontal region CT: residual calcifi- cation deep at level of basal ganglia Frequent bursts of generalized polyspike and wave, more focal discharges Generalized/ multifocal 
42, L CPS: stares, dribbles. Nocturnal GTC. Global developmental delay. Severe behavioural disturbance CT L mesial temporal calcified lesion Diffuse slow, generalized spike and wave, L temporal spikes Seizure-free for 4 months. CPS: goes quiet, abrupt version of head to R and generalization MRI: extensive L temporal lobectomy with no residual lesion. Gliotic change extending to L frontal operculum, probably postop. change Multifocal discharges max. over left frontal region. Gen. polyspike and wave in sleep. Ictal: L frontal slowing at clinical onset Ipsilateral frontal 
43, L CPS: stares, arms rigid, grimaces, wanders. GTC, some preceded by jerking of R arm. Normal development but special school from age 10. FSIQ 70  CT: left hemiatrophy L and R frontotemporal independent discharges. Ictal: (FO) L mesial temporal onset (CPS) CPS: extends R arm, grimacing, automatisms, wandering +/– GTC. Isolated GTCs. Drop attacks MRI: extensive L temporal lobectomy, gliotic changes posterior margins of resection. Diffuse loss of parenchyma of L hemisphere with signal change in white matter Diffuse slowing; multi-focal spikes. Prominent L centrotemporal spike and wave. Ictal: (FO) CPS: L frontosylvian at onset Ipsilateral frontal 
44, L CPS: high-pitched noise, aphasia, generalization Amytal: L dominant. Normal intelligence MRI: L temporal mass extending to suprasylvian region and deeply to internal capsule L temporal background abnormality, L anterior temporal spikes CPS unchanged; some reduction in generalized attacks MRI: residual tumour above and posterior to resection with deep extension to L trigone. Superior temporal gyrus infiltrated L temporosylvian spikes. Ictal: CPS recorded with no change in surface EEG  
Case no., side of surgery Preoperative clinical Preoperative imaging Preoperative EEG Postoperative clinical Postoperative imaging Postoperative EEG Electroclinical classification 
FSIQ = full scale IQ; LD = learning disability; NES = non-epileptic seizure. For other abbreviations see footnote to Table 1
35, R CPS: stares, automatisms, version to L. VIQ 96, PIQ 69. Verbal and visuospatial memory impaired CT R posterior temporal lesion involving hippocampus posteriorly Interictal: gen. spike and wave > on R, R post- temporal slow, R post- temporal spike and wave in sleep. Ictal (FO): rhythmic discharge R FO electrode (i) Frequent NES: hyperventilation and flailing of arms. (ii) Less frequent CPS: stares, rapid version to L +/– generalization Complete resection back beyond the fornix and to level of posterior horn of R lateral ventricle. No residual tumour Inter-ictal: R post- temporal spikes. Ictal: R frontocentral rhythmic theta 10 s after clinical onset Ipsilateral posterior temporal NES 
36, L CPS: stares, arms flex, head to R, posturing R arm. Sec. GTC seizures. VIQ 93, PIQ 80 CT normal Variable spiking L ant., mid and posterior temporal. FO telemetry: 3/5 seizures L FO onset, 2 similar attacks unclear onset Frequent, mainly nocturnal, seizures: stares, rapid version of head to R and elevation of R arm before generalization MRI: extensive L temporal lobectomy with posterior temporal mass lesion. Mild diffuse L hemisphere atrophy. Signal change L temporo-occipital region Ictal: rhythmic theta L central, spreading to L frontal region. Subdural mat recording: left infero- posterior temporal onset Ipsilateral posterior temporal 
37, L CPS (i) nocturnal jerking R limbs, version of head to R; (ii) fear, aphasia, upper limb automatisms. FSIQ 83, VIQ 74, PIQ 104 CT/angiography: L anterior temporal avascular lesion Diffuse slow. Gen. polyspike and wave. High-amplitude spikes L posterior temporal Frequent nocturnal GTC preceded by version of head to R and jerking of R limbs MRI: extensive L temporal resection, no residual tumour. Gliotic changes posterior to resection margin. Diffuse L hemisphere Frequent spike-wave discharges L mid and posterior temporal electrodes. Gen. symmetrical polyspike and wave atrophy Ipsilateral posterior temporal 
38, R CPS: pale, swallows, talks nonsense, runs. Hyperactive, disruptive, aggressive. Normal intelligence MRI signal abnormality R mesial temporal lobe extending to trigone of R lateral ventricle Slow waves and frequent spike and wave R anterior temporal region (i) Frequent nocturnal seizures: body twists to R, head to L with extension of L arm and flexion of R arm; (ii) confusion, runs MRI: complete resection of R temporal lesion. Simplification of R frontal gyral pattern. L mesial temporal structures normal Ictal: (i) onset anterior contacts of temporo- central mat electrode and proximal contacts of R frontal strip electrode; (ii) rhythmic spikes left anterior temporal Bilateral independent (ipsilateral frontal and contralateral temporal 
39, L Seizures: (i) vacant spells; (ii) drop attacks; (iii) dystonic jerking L limbs; (iv) GTC. FSIQ 60, VIQ 41, PIQ 68. Behavioural problems, possible psychosis CT: L mesial temporal calcified lesion extending to basal ganglia and cerebral peduncle temporal + sylvian spikes Bilateral, independent spikes, generalized slowing + gross L temporal slowing with high-amplitude L Some reduction in seizure frequency, multiple seizure types and psychological problems persist MRI: extensive L temporal lobectomy; no signs of residual tumour at resection margins. Residual tumour deeply at level of basal ganglia L temporal slowing. L suprasylvian spike and wave. Independent R-sided discharges. General discharges in sleep Generalized/ multifocal 
40, L CPS: pale, loss of awareness. GTC. Normal development to age 5, then special school. Age 5, FSIQ 78; age 14, VIQ 45, PIQ 58 CT: L mesial temporal mass extending deep to insula and to basal ganglia Diffuse slow. Multifocal and generalized spikes. Repeated discharges L posterior temporal and parietal Frequent nocturnal generalized seizures with no focal features MRI: residual tumour around margin of left temporal lobectomy, esp. deeply in the L hemisphere and the upper brainstem and thalamus Diffuse background abnormality; multifocal and general spike and wave. Focal L parasagittal discharges Generalized/ multifocal 
41, R Drop attacks. CPS: stares, R facial distortion. GTC. Normal birth, developmental regression from 4 years. Severe LD, IQ < 30 CT: calcified lesion mesial R temporal lobe R and L hemisphere Interictal EEG: bursts of general polyspike and wave, independent Severe cognitive d/sc. Carotid Amytal: suggested independence of L-sided discharges No change; continues with frequent GTC, CPS and drop attacks. hemisphere atrophy. impairment MRI: complete removal of R temporal tumour, probable diffuse R frontal region CT: residual calcifi- cation deep at level of basal ganglia Frequent bursts of generalized polyspike and wave, more focal discharges Generalized/ multifocal 
42, L CPS: stares, dribbles. Nocturnal GTC. Global developmental delay. Severe behavioural disturbance CT L mesial temporal calcified lesion Diffuse slow, generalized spike and wave, L temporal spikes Seizure-free for 4 months. CPS: goes quiet, abrupt version of head to R and generalization MRI: extensive L temporal lobectomy with no residual lesion. Gliotic change extending to L frontal operculum, probably postop. change Multifocal discharges max. over left frontal region. Gen. polyspike and wave in sleep. Ictal: L frontal slowing at clinical onset Ipsilateral frontal 
43, L CPS: stares, arms rigid, grimaces, wanders. GTC, some preceded by jerking of R arm. Normal development but special school from age 10. FSIQ 70  CT: left hemiatrophy L and R frontotemporal independent discharges. Ictal: (FO) L mesial temporal onset (CPS) CPS: extends R arm, grimacing, automatisms, wandering +/– GTC. Isolated GTCs. Drop attacks MRI: extensive L temporal lobectomy, gliotic changes posterior margins of resection. Diffuse loss of parenchyma of L hemisphere with signal change in white matter Diffuse slowing; multi-focal spikes. Prominent L centrotemporal spike and wave. Ictal: (FO) CPS: L frontosylvian at onset Ipsilateral frontal 
44, L CPS: high-pitched noise, aphasia, generalization Amytal: L dominant. Normal intelligence MRI: L temporal mass extending to suprasylvian region and deeply to internal capsule L temporal background abnormality, L anterior temporal spikes CPS unchanged; some reduction in generalized attacks MRI: residual tumour above and posterior to resection with deep extension to L trigone. Superior temporal gyrus infiltrated L temporosylvian spikes. Ictal: CPS recorded with no change in surface EEG  
Table 5

Electroclinical classification of postoperative seizures

Pathology (no. of cases) Ipsilateral Contralateral Generalized/multifocal NES Uncertain 
 Temporal Frontal Occipital Centro-sylvian Hemisphere Temporal Frontal    
*Three cases had an additional seizure type; two cases had an additional seizure type; also had ipsilateral temporal seizures. 
MTS (20) 12 – – – – 
DNT (10) – – – – 1 – 
NSP (14) – 
Total (44) 20* 6 8 1 
Pathology (no. of cases) Ipsilateral Contralateral Generalized/multifocal NES Uncertain 
 Temporal Frontal Occipital Centro-sylvian Hemisphere Temporal Frontal    
*Three cases had an additional seizure type; two cases had an additional seizure type; also had ipsilateral temporal seizures. 
MTS (20) 12 – – – – 
DNT (10) – – – – 1 – 
NSP (14) – 
Total (44) 20* 6 8 1 

References

Amaral DG. Amygdalohippocampal and amygdalocortical projections in the primate brain. In: Schwarcz R, Ben-Ari Y, editors. Excitatory amino acids and epilepsy. New York: Plenum Press; 1986. p. 3–17.
Awad IA, Nayel MH, Luders H. Second operation after the failure of previous resection for epilepsy.
Neurosurgery
 
1991
;
28
:
510
–8.
Barkovich AJ, Rowley HA, Andermann F. MR in partial epilepsy: value of high-resolution volumetric techniques.
AJNR Am J Neuroradiol
 
1995
;
16
:
339
–43.
Bastos AC, Comeau RM, Andermann F, Melanson D, Cendes F, Dubeau F, et al. Diagnosis of subtle focal dysplastic lesions. Curvilinear reformatting from three-dimensional magnetic resonance imaging.
Ann Neurol
 
1999
;
46
:
88
–94.
Berkovic SF, McIntosh AM, Kalnins RM, Jackson GD, Fabinyi GC, Brazenor GA, et al. Preoperative MRI predicts outcome of temporal lobectomy: an actuarial analysis.
Neurology
 
1995
;
45
:
1358
–63.
Bruton CJ. The neuropathology of temporal lobe epilepsy. Institute of Psychiatry, Maudsley Monograph No. 31. Oxford: Oxford University Press; 1988. p. 74–5.
Burgerman RS, Sperling MR, French JA, Saykin AJ, O'Connor MJ. Comparison of mesial versus neocortical onset temporal lobe seizures: neurodiagnostic findings and surgical outcome.
Epilepsia
 
1995
;
36
:
662
–70.
Cascino GD, Andermann F, Berkovic SF, Kuzniecky RI, Sharbrough FW, Keene DL, et al. Gelastic seizures and hypothalamic hamartomas: evaluation of patients undergoing chronic intracranial EEG monitoring and outcome of surgical treatment.
Neurology
 
1993
;
43
:
747
–50.
Cendes F, Cook MJ, Watson C, Andermann F, Fish DR, Shorvon SD, et al. Frequency and characteristics of dual pathology in patients with lesional epilepsy.
Neurology
 
1995
;
45
:
2058
–64.
Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for classification of epilepsies and epileptic syndromes.
Epilepsia
 
1985
;
26
(Suppl 3),
268
–78.
Daumas-Duport C, Scheithauer BW, Chodkiewicz JP, Laws ER Jr, Vedrenne C. Dysembryoplastic neuroepithelial tumor: a surgically curable tumor of young patients with intractable partial seizures.
Neurosurgery
 
1988
;
23
:
545
–56.
Duchowny M, Jayakar P, Resnick T, Levin B, Alvarez L. Posterior temporal epilepsy: electroclinical features.
Ann Neurol
 
1994
;
35
:
427
–31.
Duncan JS, Sagar HJ. Seizure characteristics, pathology, and outcome after temporal lobectomy.
Neurology
 
1987
;
37
:
405
–9.
Elwes RD, Dunn G, Binnie CD, Polkey CE. Outcome following resective surgery for temporal lobe epilepsy: a prospective follow-up study of 102 consecutive cases.
J Neurol Neurosurg Psychiatry
 
1991
;
54
:
949
–52.
Engel J Jr, Van Ness P, Rasmussen T. Outcome with respect to epileptic seizures. In: Engel J Jr, ed. Surgical treatment of the epilepsies 2nd ed. New York: Raven Press, 1993. p. 609–21.
Fakhoury T, Abou-Khalil B. Association of ipsilateral head turning and dystonia in temporal lobe seizures.
Epilepsia
 
1995
;
36
:
1065
–70.
Fakhoury T, Abou-Khalil B, Peguero E. Differentiating clinical features of right and left temporal lobe seizures.
Epilepsia
 
1994
;
35
:
1038
–44.
Falconer MA, Serafetinides EA, Corsellis JAN. Etiology and pathogenesis of temporal lobe epilepsy.
Arch Neurol
 
1964
;
10
:
233
–48.
Fernandez Torre JL, Alarcon G, Binnie CD, Seoane JJ, Juler J, Guy CN, et al. Generation of scalp discharges in temporal lobe epilepsy as suggested by intraoperative electrocorticographic recordings.
J Neurol Neurosurg Psychiatry
 
1999
;
67
:
51
–8.
French JA, Williamson PD, Thadani VM, Darcey TM, Mattson RH, Spencer SS, et al. Characteristics of medial temporal lobe epilepsy: 1. Results of history and physical examination.
Ann Neurol
 
1993
;
34
:
774
–80.
Germano IM, Poulin N, Olivier A. Reoperation for recurrent temporal lobe epilepsy.
J Neurosurg
 
1994
;
81
:
31
–6.
Gloor P. Experiential phenomena of temporal lobe epilepsy. [Review].
Brain
 
1990
;
113
:
1673
–94.
Hennessy MJ, Langan Y, Elwes RD, Binnie CD, Polkey CE, Nashef L. A study of mortality after temporal lobe epilepsy surgery.
Neurology
 
1999
;
53
:
1276
–83.
Honavar M, Janota I, Polkey CE. Neuropathological findings in surgery for epilepsy. Four hundred and thirty resections for treatment of chronic epilepsy: an 18 year experience.
Bol Epilepsia
 
1997
;
2
:
14
–28.
Jack CR Jr, Sharbrough FW, Twomey CK, Cascino GD, Hirschorn KA, Marsh WR, et al. Temporal lobe seizures: lateralization with MR volume measurements of the hippocampal formation.
Radiology
 
1990
;
175
:
423
–9.
Jayakar P, Duchowny M, Resnick T, Alvarez L. Ictal head deviation: lateralizing significance of the pattern of head movement.
Neurology
 
1992
;
42
:
1989
–92.
King D, Spencer SS, McCarthy G, Luby M, Spencer DD. Bilateral hippocampal atrophy in medial temporal lobe epilepsy.
Epilepsia
 
1995
;
36
:
905
–10.
Kirkpatrick PJ, Honavar M, Janota I, Polkey CE. Control of temporal lobe epilepsy following en bloc resection of low grade tumors.
J Neurosurg
 
1993
;
78
:
19
–25.
Koerner M, Laxer KD. Ictal speech, postictal language dysfunction, and seizure lateralization.
Neurology
 
1988
;
38
:
634
–6.
Kotagal P. Seizure symptomatology of temporal lobe origin. In: Luders H, editor. Epilepsy surgery. New York: Raven Press; 1991. p. 143–55.
Kotagal P, Luders H, Morris HH, Dinner DS, Wyllie E, Godoy J, Rothner AD. Dystonic posturing in complex partial seizures of temporal lobe onset; a new lateralizing sign.
Neurology
 
1989
;
39
:
196
–201.
Kuzniecky RI. Magnetic resonance imaging in developmental disorders of the cerebral cortex. [Review].
Epilepsia
 
1994
;
35
Suppl 6:
S44
–56.
Kuzniecky R, Guthrie B, Mountz J, Bebin M, Faught E, Gilliam F, et al. Intrinsic epileptogenesis of hypothalamic hamartomas in gelastic epilepsy.
Ann Neurol
 
1997
;
42
:
60
–7.
Lee JW, Reutens DC, Dubeau DC, Evans A, Andermann F. Morphometry in temporal lobe epilepsy.
Magn Reson Imaging
 
1995
;
13
:
1073
–80.
Lencz T, McCarthy G, Bronen RA, Scott TM, Inserni JA, Sass KJ, et al. Quantitative magnetic resonance imaging in temporal lobe epilepsy: relationship to neuropathology and neuropsychological function.
Ann Neurol
 
1992
;
31
:
629
–37.
Li LM, Dubeau F, Andermann F, Fish DR, Watson C, Cascino GD, et al. Periventricular nodular heterotopia and intractable temporal lobe epilepsy: poor outcome after temporal lobe resection.
Ann Neurol
 
1997
;
41
:
662
–8.
Li LM, Cendes F, Andermann F, Watson C, Fish DR, Cook MJ, et al. Surgical outcome in patients with epilepsy and dual pathology.
Brain
 
1999
;
122
:
799
–805.
Ludwig BI, Marsan CA. Clinical ictal patterns in epileptic patients with occipital electroencephalographic foci.
Neurology
 
1975
;
25
:
463
–71.
Margerison JH, Corsellis JA. Epilepsy and the temporal lobes.
Brain
 
1966
;
89
:
499
–530.
Marsh L, Morrell MJ, Shear PK, Sullivan EV, Freeman H, Marie A, et al. Cortical and hippocampal volume deficits in temporal lobe epilepsy.
Epilepsia
 
1997
;
38
:
576
–87.
Matsuo F, Knott JR. Focal positive spikes in electroencephalography.
Electroencephalogr Clin Neurophysiol
 
1977
;
42
:
15
–25.
Meencke HJ, Veith G. Hippocampal sclerosis in epilepsy. In Lüders H, ed. Epilepsy surgery. New York: Raven Press 1991. p. 705–15.
Morrell F. Secondary epileptogenesis in man.
Arch Neurol
 
1985
;
42
:
318
–35.
Munari C, Kahane P, Francione S, Hoffmann D, Tassi L, Cusmai R. Role of the hypothalamic hamartoma in the genesis of gelastic fits.
Electroencephalogr Clin Neurophysiol
 
1995
;
95
:
154
–60.
Nakasato N, Levesque MF, Babb TL. Seizure outcome following standard temporal lobectomy: correlation with hippocampal neuron loss and extrahippocampal pathology.
J Neurosurg
 
1992
;
77
:
194
–200.
Nayel MH, Awad IA, Luders H. Extent of mesiobasal resection determines outcome after temporal lobectomy for intractable complex partial seizures.
Neurosurgery
 
1991
;
29
:
55
–61.
Newton MR, Berkovic SF, Austin MC, Reutens DC, McKay WJ, Bladin PF. Dystonia, clinical lateralization, and regional blood flow changes in temporal lobe seizures.
Neurology
 
1992
;
42
:
371
–7.
Pacia SV, Devinsky O, Perrine K, Ravdin L, Luciano D, Vazquez B, et al. Clinical features of neocortical temporal lobe epilepsy.
Ann Neurol
 
1996
;
40
:
724
–30.
Penfield W, Jasper H. Functional localization in the cerebral cortex. In: Penfield W, Jasper H. Epilepsy and the functional anatomy of the human brain. London: J. and A. Churchill; 1954. p. 113–6.
Polkey CE. Neurosurgery. In Laidlaw J, Richens A, Oxley J, editors. A textbook of epilepsy. 3rd ed. Edinburgh: Churchill Livingstone; 1988. p. 493.
Prayson RA, Estes ML, Morris HH. Coexistence of neoplasia and cortical dysplasia in patients presenting with seizures.
Epilepsia
 
1993
;
34
:
609
–15.
Prayson RA, Morris HH, Estes ML, Comair YG. Dysembryoplastic neuroepithelial tumor: a clinicopathologic and immunohistochemical study of 11 tumors including MIBI immunoreactivity.
Clin Neuropathol
 
1996
;
15
:
47
–53.
Privitera MD, Morris GL, Gilliam F. Postictal language assessment and lateralization of complex partial seizures.
Ann Neurol
 
1991
;
30
:
391
–6.
Quigg M, Bertram EH, Jackson T, Laws E. Volumetric magnetic resonance imaging evidence of bilateral hippocampal atrophy in mesial temporal lobe epilepsy.
Epilepsia
 
1997
;
38
:
588
–94.
Raymond AA, Fish DR, Stevens JM, Cook MJ, Sisodiya SM, Shorvon SD. Association of hippocampal sclerosis with cortical dysgenesis in patients with epilepsy.
Neurology
 
1994
;
44
:
1841
–5.
Raymond AA, Halpin SF, Alsanjari N, Cook MJ, Kitchen ND, Fish DR, et al. Dysembryoplastic neuroepithelial tumor. Features in 16 patients.
Brain
 
1994
;
117
:
461
–75.
Richardson MP, Koepp MJ, Brooks DJ, Duncan JS. 11C-flumazenil PET in neocortical epilepsy.
Neurology
 
1998
;
51
:
485
–92.
Sagar HJ, Oxbury JM. Hippocampal neuron loss in temporal lobe epilepsy: correlation with early childhood convulsions.
Ann Neurol
 
1987
;
22
:
334
–40.
Salanova V, Andermann F, Olivier A, Rasmussen T, Quesney LF. Occipital lobe epilepsy: electroclinical manifestations, electrocorticography, cortical stimulation and outcome in 42 patients treated between 1930 and 1991.
Brain
 
1992
;
115
:
1655
–80.
Sano K, Malamud N. Clinical significance of sclerosis of the cornu ammonis.
Arch Neurol Psychiatr
 
1953
;
70
:
40
–53.
Sisodiya SM, Free SL. Disproportion of cerebral surface areas and volumes in cerebral dysgenesis. MRI-based evidence for connectional abnormalities.
Brain
 
1997
;
120
:
271
–81.
Sisodiya SM, Moran N, Free SL, Kitchen ND, Stevens JM, Harkness WF, et al. Correlation of widespread preoperative magnetic resonance imaging changes with unsuccessful surgery for hippocampal sclerosis.
Ann Neurol
 
1997
;
41
:
490
–6.
Spencer DD, Spencer SS, Mattson RH, Williamson PD, Novelly RA. Access to the posterior medial temporal lobe structures in the surgical treatment of temporal lobe epilepsy.
Neurosurgery
 
1984
;
15
:
667
–71.
Spencer SS. Long-term outcome after epilepsy surgery. [Review].
Epilepsia
 
1996
;
37
:
807
–13.
Spencer SS, Williamson PD, Spencer DD, Mattson RH. Human hippocampal seizure spread studied by depth and subdural recording: the hippocampal commissure.
Epilepsia
 
1987
;
28
:
479
–89.
Taratuto AL, Pomata H, Sevlever G, Gallo G, Monges J. Dysembryoplastic neuroepithelial tumor: morphological, immunocytochemical, and deoxyribonucleic acid analyses in a pediatric series.
Neurosurgery
 
1995
;
36
:
474
–81.
Thadani VM, Williamson PD, Berger R, Spencer SS, Spencer DD, Novelly RA, et al. Successful epilepsy surgery without intracranial EEG recording: criteria for patient selection.
Epilepsia
 
1995
;
36
:
7
–15.
Van Paesschen W, Connelly A, King MD, Jackson GD, Duncan JS. The spectrum of hippocampal sclerosis: a quantitative magnetic resonance imaging study.
Ann Neurol
 
1997
;
41
:
41
–51.
Williamson PD, Thadani VM, Darcey TM, Spencer DD, Spencer SS, Mattson RH. Occipital lobe epilepsy: clinical characteristics, seizure spread patterns, and results of surgery. [Review].
Ann Neurol
 
1992
;
31
:
3
–13.
Williamson PD, Boon PA, Thadani VM, Darcey TM, Spencer DD, Spencer SS, et al. Parietal lobe epilepsy: diagnostic considerations and results of surgery.
Ann Neurol
 
1992
;
31
:
193
–201.
Williamson PD, Thadani VM, French JA, Darcey TM, Mattson RH, Spencer SS, et al. Medial temporal lobe epilepsy: videotape analysis of objective clinical seizure characteristics.
Epilepsia
 
1998
;
39
:
1182
–8.
Wyler AR, Hermann BP, Richey ET. Results of reoperation for failed epilepsy surgery.
J Neurosurg
 
1989
;
71
:
815
–9.
Wyler AR, Hermann BP, Somes G. Extent of medial temporal resection on outcome from anterior temporal lobectomy: a randomized prospective study.
Neurosurgery
 
1995
;
37
:
982
–90.
Wyllie E, Luders H, Morris HH, Lesser RP, Dinner DS. The lateralizing significance of versive head and eye movements during epileptic seizures.
Neurology
 
1986
;
36
:
606
–11.
Yasargil MG, Teddy PJ, Roth P. Selective amygdalo-hippocampectomy: operative anatomy and surgical technique.
Adv Tech Stand Neurosurg
 
1985
;
12
:
93
–123.
Yen D, Su MS, Yiu CH, Shih YH, Kwan SY, Tsai CP, et al. Ictal speech manifestations in temporal lobe epilepsy: a video-EEG study.
Epilepsia
 
1996
;
37
:
45
–9.
Zentner J, Hufnagel A, Wolf HK, Ostertun B, Behrens E, Campos MG, et al. Surgical treatment of temporal lobe epilepsy: clinical, radiological, and histopathological findings in 178 patients.
J Neurol Neurosurg Psychiatry
 
1995
;
58
:
666
–73.